backend: make it clear that we passed 'fd' to the new thread
[fio.git] / HOWTO
CommitLineData
f80dba8d
MT
1How fio works
2-------------
3
4The first step in getting fio to simulate a desired I/O workload, is writing a
5job file describing that specific setup. A job file may contain any number of
6threads and/or files -- the typical contents of the job file is a *global*
7section defining shared parameters, and one or more job sections describing the
8jobs involved. When run, fio parses this file and sets everything up as
9described. If we break down a job from top to bottom, it contains the following
10basic parameters:
11
12`I/O type`_
13
14 Defines the I/O pattern issued to the file(s). We may only be reading
15 sequentially from this file(s), or we may be writing randomly. Or even
16 mixing reads and writes, sequentially or randomly.
17 Should we be doing buffered I/O, or direct/raw I/O?
18
19`Block size`_
20
21 In how large chunks are we issuing I/O? This may be a single value,
22 or it may describe a range of block sizes.
23
24`I/O size`_
25
26 How much data are we going to be reading/writing.
27
28`I/O engine`_
29
30 How do we issue I/O? We could be memory mapping the file, we could be
31 using regular read/write, we could be using splice, async I/O, or even
32 SG (SCSI generic sg).
33
34`I/O depth`_
35
36 If the I/O engine is async, how large a queuing depth do we want to
37 maintain?
38
39
40`Target file/device`_
41
42 How many files are we spreading the workload over.
43
44`Threads, processes and job synchronization`_
45
46 How many threads or processes should we spread this workload over.
47
48The above are the basic parameters defined for a workload, in addition there's a
49multitude of parameters that modify other aspects of how this job behaves.
50
51
52Command line options
53--------------------
54
55.. option:: --debug=type
56
f50fbdda 57 Enable verbose tracing `type` of various fio actions. May be ``all`` for all types
b034c0dd
TK
58 or individual types separated by a comma (e.g. ``--debug=file,mem`` will
59 enable file and memory debugging). Currently, additional logging is
60 available for:
f80dba8d 61
b034c0dd 62 *process*
f80dba8d 63 Dump info related to processes.
b034c0dd 64 *file*
f80dba8d 65 Dump info related to file actions.
b034c0dd 66 *io*
f80dba8d 67 Dump info related to I/O queuing.
b034c0dd 68 *mem*
f80dba8d 69 Dump info related to memory allocations.
b034c0dd 70 *blktrace*
f80dba8d 71 Dump info related to blktrace setup.
b034c0dd 72 *verify*
f80dba8d 73 Dump info related to I/O verification.
b034c0dd 74 *all*
f80dba8d 75 Enable all debug options.
b034c0dd 76 *random*
f80dba8d 77 Dump info related to random offset generation.
b034c0dd 78 *parse*
f80dba8d 79 Dump info related to option matching and parsing.
b034c0dd 80 *diskutil*
f80dba8d 81 Dump info related to disk utilization updates.
b034c0dd 82 *job:x*
f80dba8d 83 Dump info only related to job number x.
b034c0dd 84 *mutex*
f80dba8d 85 Dump info only related to mutex up/down ops.
b034c0dd 86 *profile*
f80dba8d 87 Dump info related to profile extensions.
b034c0dd 88 *time*
f80dba8d 89 Dump info related to internal time keeping.
b034c0dd 90 *net*
f80dba8d 91 Dump info related to networking connections.
b034c0dd 92 *rate*
f80dba8d 93 Dump info related to I/O rate switching.
b034c0dd 94 *compress*
f80dba8d 95 Dump info related to log compress/decompress.
b034c0dd 96 *?* or *help*
f80dba8d
MT
97 Show available debug options.
98
99.. option:: --parse-only
100
25cd4b95 101 Parse options only, don't start any I/O.
f80dba8d
MT
102
103.. option:: --output=filename
104
105 Write output to file `filename`.
106
f50fbdda 107.. option:: --output-format=format
b8f7e412 108
f50fbdda 109 Set the reporting `format` to `normal`, `terse`, `json`, or `json+`. Multiple
b8f7e412
TK
110 formats can be selected, separated by a comma. `terse` is a CSV based
111 format. `json+` is like `json`, except it adds a full dump of the latency
112 buckets.
113
f80dba8d
MT
114.. option:: --bandwidth-log
115
116 Generate aggregate bandwidth logs.
117
118.. option:: --minimal
119
120 Print statistics in a terse, semicolon-delimited format.
121
122.. option:: --append-terse
123
b034c0dd
TK
124 Print statistics in selected mode AND terse, semicolon-delimited format.
125 **Deprecated**, use :option:`--output-format` instead to select multiple
126 formats.
f80dba8d 127
f50fbdda 128.. option:: --terse-version=version
f80dba8d 129
f50fbdda 130 Set terse `version` output format (default 3, or 2 or 4 or 5).
f80dba8d
MT
131
132.. option:: --version
133
b8f7e412 134 Print version information and exit.
f80dba8d
MT
135
136.. option:: --help
137
113f0e7c 138 Print a summary of the command line options and exit.
f80dba8d
MT
139
140.. option:: --cpuclock-test
141
142 Perform test and validation of internal CPU clock.
143
113f0e7c 144.. option:: --crctest=[test]
f80dba8d 145
b034c0dd
TK
146 Test the speed of the built-in checksumming functions. If no argument is
147 given, all of them are tested. Alternatively, a comma separated list can
148 be passed, in which case the given ones are tested.
f80dba8d
MT
149
150.. option:: --cmdhelp=command
151
152 Print help information for `command`. May be ``all`` for all commands.
153
154.. option:: --enghelp=[ioengine[,command]]
155
f50fbdda
TK
156 List all commands defined by `ioengine`, or print help for `command`
157 defined by `ioengine`. If no `ioengine` is given, list all
b034c0dd 158 available ioengines.
f80dba8d
MT
159
160.. option:: --showcmd=jobfile
161
b8f7e412 162 Convert `jobfile` to a set of command-line options.
f80dba8d
MT
163
164.. option:: --readonly
165
b034c0dd
TK
166 Turn on safety read-only checks, preventing writes. The ``--readonly``
167 option is an extra safety guard to prevent users from accidentally starting
168 a write workload when that is not desired. Fio will only write if
169 `rw=write/randwrite/rw/randrw` is given. This extra safety net can be used
170 as an extra precaution as ``--readonly`` will also enable a write check in
171 the I/O engine core to prevent writes due to unknown user space bug(s).
f80dba8d
MT
172
173.. option:: --eta=when
174
b8f7e412
TK
175 Specifies when real-time ETA estimate should be printed. `when` may be
176 `always`, `never` or `auto`.
f80dba8d
MT
177
178.. option:: --eta-newline=time
179
947e0fe0
SW
180 Force a new line for every `time` period passed. When the unit is omitted,
181 the value is interpreted in seconds.
f80dba8d
MT
182
183.. option:: --status-interval=time
184
aa6cb459
VF
185 Force a full status dump of cumulative (from job start) values at `time`
186 intervals. This option does *not* provide per-period measurements. So
187 values such as bandwidth are running averages. When the time unit is omitted,
188 `time` is interpreted in seconds.
f80dba8d
MT
189
190.. option:: --section=name
191
b034c0dd
TK
192 Only run specified section `name` in job file. Multiple sections can be specified.
193 The ``--section`` option allows one to combine related jobs into one file.
194 E.g. one job file could define light, moderate, and heavy sections. Tell
195 fio to run only the "heavy" section by giving ``--section=heavy``
196 command line option. One can also specify the "write" operations in one
197 section and "verify" operation in another section. The ``--section`` option
198 only applies to job sections. The reserved *global* section is always
199 parsed and used.
f80dba8d
MT
200
201.. option:: --alloc-size=kb
202
b034c0dd
TK
203 Set the internal smalloc pool size to `kb` in KiB. The
204 ``--alloc-size`` switch allows one to use a larger pool size for smalloc.
205 If running large jobs with randommap enabled, fio can run out of memory.
206 Smalloc is an internal allocator for shared structures from a fixed size
207 memory pool and can grow to 16 pools. The pool size defaults to 16MiB.
f80dba8d 208
b034c0dd
TK
209 NOTE: While running :file:`.fio_smalloc.*` backing store files are visible
210 in :file:`/tmp`.
f80dba8d
MT
211
212.. option:: --warnings-fatal
213
b034c0dd
TK
214 All fio parser warnings are fatal, causing fio to exit with an
215 error.
f80dba8d
MT
216
217.. option:: --max-jobs=nr
218
f50fbdda 219 Set the maximum number of threads/processes to support to `nr`.
818322cc 220 NOTE: On Linux, it may be necessary to increase the shared-memory
71aa48eb 221 limit (:file:`/proc/sys/kernel/shmmax`) if fio runs into errors while
818322cc 222 creating jobs.
f80dba8d
MT
223
224.. option:: --server=args
225
b034c0dd
TK
226 Start a backend server, with `args` specifying what to listen to.
227 See `Client/Server`_ section.
f80dba8d
MT
228
229.. option:: --daemonize=pidfile
230
b034c0dd 231 Background a fio server, writing the pid to the given `pidfile` file.
f80dba8d
MT
232
233.. option:: --client=hostname
234
f50fbdda 235 Instead of running the jobs locally, send and run them on the given `hostname`
71aa48eb 236 or set of `hostname`\s. See `Client/Server`_ section.
f80dba8d
MT
237
238.. option:: --remote-config=file
239
f50fbdda 240 Tell fio server to load this local `file`.
f80dba8d
MT
241
242.. option:: --idle-prof=option
243
b8f7e412 244 Report CPU idleness. `option` is one of the following:
113f0e7c
SW
245
246 **calibrate**
247 Run unit work calibration only and exit.
248
249 **system**
250 Show aggregate system idleness and unit work.
251
252 **percpu**
253 As **system** but also show per CPU idleness.
f80dba8d
MT
254
255.. option:: --inflate-log=log
256
f50fbdda 257 Inflate and output compressed `log`.
f80dba8d
MT
258
259.. option:: --trigger-file=file
260
f50fbdda 261 Execute trigger command when `file` exists.
f80dba8d 262
f50fbdda 263.. option:: --trigger-timeout=time
f80dba8d 264
f50fbdda 265 Execute trigger at this `time`.
f80dba8d 266
f50fbdda 267.. option:: --trigger=command
f80dba8d 268
f50fbdda 269 Set this `command` as local trigger.
f80dba8d 270
f50fbdda 271.. option:: --trigger-remote=command
f80dba8d 272
f50fbdda 273 Set this `command` as remote trigger.
f80dba8d
MT
274
275.. option:: --aux-path=path
276
f50fbdda 277 Use this `path` for fio state generated files.
f80dba8d
MT
278
279Any parameters following the options will be assumed to be job files, unless
280they match a job file parameter. Multiple job files can be listed and each job
281file will be regarded as a separate group. Fio will :option:`stonewall`
282execution between each group.
283
284
285Job file format
286---------------
287
288As previously described, fio accepts one or more job files describing what it is
289supposed to do. The job file format is the classic ini file, where the names
c60ebc45 290enclosed in [] brackets define the job name. You are free to use any ASCII name
f80dba8d
MT
291you want, except *global* which has special meaning. Following the job name is
292a sequence of zero or more parameters, one per line, that define the behavior of
293the job. If the first character in a line is a ';' or a '#', the entire line is
294discarded as a comment.
295
296A *global* section sets defaults for the jobs described in that file. A job may
297override a *global* section parameter, and a job file may even have several
298*global* sections if so desired. A job is only affected by a *global* section
299residing above it.
300
f50fbdda
TK
301The :option:`--cmdhelp` option also lists all options. If used with a `command`
302argument, :option:`--cmdhelp` will detail the given `command`.
f80dba8d
MT
303
304See the `examples/` directory for inspiration on how to write job files. Note
305the copyright and license requirements currently apply to `examples/` files.
306
307So let's look at a really simple job file that defines two processes, each
308randomly reading from a 128MiB file:
309
310.. code-block:: ini
311
312 ; -- start job file --
313 [global]
314 rw=randread
315 size=128m
316
317 [job1]
318
319 [job2]
320
321 ; -- end job file --
322
323As you can see, the job file sections themselves are empty as all the described
324parameters are shared. As no :option:`filename` option is given, fio makes up a
325`filename` for each of the jobs as it sees fit. On the command line, this job
326would look as follows::
327
328$ fio --name=global --rw=randread --size=128m --name=job1 --name=job2
329
330
331Let's look at an example that has a number of processes writing randomly to
332files:
333
334.. code-block:: ini
335
336 ; -- start job file --
337 [random-writers]
338 ioengine=libaio
339 iodepth=4
340 rw=randwrite
341 bs=32k
342 direct=0
343 size=64m
344 numjobs=4
345 ; -- end job file --
346
347Here we have no *global* section, as we only have one job defined anyway. We
348want to use async I/O here, with a depth of 4 for each file. We also increased
349the buffer size used to 32KiB and define numjobs to 4 to fork 4 identical
350jobs. The result is 4 processes each randomly writing to their own 64MiB
351file. Instead of using the above job file, you could have given the parameters
352on the command line. For this case, you would specify::
353
354$ fio --name=random-writers --ioengine=libaio --iodepth=4 --rw=randwrite --bs=32k --direct=0 --size=64m --numjobs=4
355
356When fio is utilized as a basis of any reasonably large test suite, it might be
357desirable to share a set of standardized settings across multiple job files.
358Instead of copy/pasting such settings, any section may pull in an external
359:file:`filename.fio` file with *include filename* directive, as in the following
360example::
361
362 ; -- start job file including.fio --
363 [global]
364 filename=/tmp/test
365 filesize=1m
366 include glob-include.fio
367
368 [test]
369 rw=randread
370 bs=4k
371 time_based=1
372 runtime=10
373 include test-include.fio
374 ; -- end job file including.fio --
375
376.. code-block:: ini
377
378 ; -- start job file glob-include.fio --
379 thread=1
380 group_reporting=1
381 ; -- end job file glob-include.fio --
382
383.. code-block:: ini
384
385 ; -- start job file test-include.fio --
386 ioengine=libaio
387 iodepth=4
388 ; -- end job file test-include.fio --
389
390Settings pulled into a section apply to that section only (except *global*
391section). Include directives may be nested in that any included file may contain
392further include directive(s). Include files may not contain [] sections.
393
394
395Environment variables
396~~~~~~~~~~~~~~~~~~~~~
397
398Fio also supports environment variable expansion in job files. Any sub-string of
399the form ``${VARNAME}`` as part of an option value (in other words, on the right
400of the '='), will be expanded to the value of the environment variable called
401`VARNAME`. If no such environment variable is defined, or `VARNAME` is the
402empty string, the empty string will be substituted.
403
404As an example, let's look at a sample fio invocation and job file::
405
406$ SIZE=64m NUMJOBS=4 fio jobfile.fio
407
408.. code-block:: ini
409
410 ; -- start job file --
411 [random-writers]
412 rw=randwrite
413 size=${SIZE}
414 numjobs=${NUMJOBS}
415 ; -- end job file --
416
417This will expand to the following equivalent job file at runtime:
418
419.. code-block:: ini
420
421 ; -- start job file --
422 [random-writers]
423 rw=randwrite
424 size=64m
425 numjobs=4
426 ; -- end job file --
427
428Fio ships with a few example job files, you can also look there for inspiration.
429
430Reserved keywords
431~~~~~~~~~~~~~~~~~
432
433Additionally, fio has a set of reserved keywords that will be replaced
434internally with the appropriate value. Those keywords are:
435
436**$pagesize**
437
438 The architecture page size of the running system.
439
440**$mb_memory**
441
442 Megabytes of total memory in the system.
443
444**$ncpus**
445
446 Number of online available CPUs.
447
448These can be used on the command line or in the job file, and will be
449automatically substituted with the current system values when the job is
450run. Simple math is also supported on these keywords, so you can perform actions
451like::
452
b034c0dd 453 size=8*$mb_memory
f80dba8d
MT
454
455and get that properly expanded to 8 times the size of memory in the machine.
456
457
458Job file parameters
459-------------------
460
461This section describes in details each parameter associated with a job. Some
462parameters take an option of a given type, such as an integer or a
463string. Anywhere a numeric value is required, an arithmetic expression may be
464used, provided it is surrounded by parentheses. Supported operators are:
465
466 - addition (+)
467 - subtraction (-)
468 - multiplication (*)
469 - division (/)
470 - modulus (%)
471 - exponentiation (^)
472
473For time values in expressions, units are microseconds by default. This is
474different than for time values not in expressions (not enclosed in
475parentheses). The following types are used:
476
477
478Parameter types
479~~~~~~~~~~~~~~~
480
481**str**
b034c0dd 482 String: A sequence of alphanumeric characters.
f80dba8d
MT
483
484**time**
008d0feb
SW
485 Integer with possible time suffix. Without a unit value is interpreted as
486 seconds unless otherwise specified. Accepts a suffix of 'd' for days, 'h' for
487 hours, 'm' for minutes, 's' for seconds, 'ms' (or 'msec') for milliseconds and
488 'us' (or 'usec') for microseconds. For example, use 10m for 10 minutes.
f80dba8d
MT
489
490.. _int:
491
492**int**
493 Integer. A whole number value, which may contain an integer prefix
494 and an integer suffix:
495
b034c0dd 496 [*integer prefix*] **number** [*integer suffix*]
f80dba8d
MT
497
498 The optional *integer prefix* specifies the number's base. The default
499 is decimal. *0x* specifies hexadecimal.
500
501 The optional *integer suffix* specifies the number's units, and includes an
502 optional unit prefix and an optional unit. For quantities of data, the
947e0fe0
SW
503 default unit is bytes. For quantities of time, the default unit is seconds
504 unless otherwise specified.
f80dba8d 505
9207a0cb 506 With :option:`kb_base`\=1000, fio follows international standards for unit
f80dba8d
MT
507 prefixes. To specify power-of-10 decimal values defined in the
508 International System of Units (SI):
509
eccce61a
TK
510 * *K* -- means kilo (K) or 1000
511 * *M* -- means mega (M) or 1000**2
512 * *G* -- means giga (G) or 1000**3
513 * *T* -- means tera (T) or 1000**4
514 * *P* -- means peta (P) or 1000**5
f80dba8d
MT
515
516 To specify power-of-2 binary values defined in IEC 80000-13:
517
eccce61a
TK
518 * *Ki* -- means kibi (Ki) or 1024
519 * *Mi* -- means mebi (Mi) or 1024**2
520 * *Gi* -- means gibi (Gi) or 1024**3
521 * *Ti* -- means tebi (Ti) or 1024**4
522 * *Pi* -- means pebi (Pi) or 1024**5
f80dba8d 523
9207a0cb 524 With :option:`kb_base`\=1024 (the default), the unit prefixes are opposite
f80dba8d
MT
525 from those specified in the SI and IEC 80000-13 standards to provide
526 compatibility with old scripts. For example, 4k means 4096.
527
528 For quantities of data, an optional unit of 'B' may be included
b8f7e412 529 (e.g., 'kB' is the same as 'k').
f80dba8d
MT
530
531 The *integer suffix* is not case sensitive (e.g., m/mi mean mebi/mega,
532 not milli). 'b' and 'B' both mean byte, not bit.
533
9207a0cb 534 Examples with :option:`kb_base`\=1000:
f80dba8d
MT
535
536 * *4 KiB*: 4096, 4096b, 4096B, 4ki, 4kib, 4kiB, 4Ki, 4KiB
537 * *1 MiB*: 1048576, 1mi, 1024ki
538 * *1 MB*: 1000000, 1m, 1000k
539 * *1 TiB*: 1099511627776, 1ti, 1024gi, 1048576mi
540 * *1 TB*: 1000000000, 1t, 1000m, 1000000k
541
9207a0cb 542 Examples with :option:`kb_base`\=1024 (default):
f80dba8d
MT
543
544 * *4 KiB*: 4096, 4096b, 4096B, 4k, 4kb, 4kB, 4K, 4KB
545 * *1 MiB*: 1048576, 1m, 1024k
546 * *1 MB*: 1000000, 1mi, 1000ki
547 * *1 TiB*: 1099511627776, 1t, 1024g, 1048576m
548 * *1 TB*: 1000000000, 1ti, 1000mi, 1000000ki
549
550 To specify times (units are not case sensitive):
551
552 * *D* -- means days
553 * *H* -- means hours
4502cb42 554 * *M* -- means minutes
f80dba8d
MT
555 * *s* -- or sec means seconds (default)
556 * *ms* -- or *msec* means milliseconds
557 * *us* -- or *usec* means microseconds
558
559 If the option accepts an upper and lower range, use a colon ':' or
560 minus '-' to separate such values. See :ref:`irange <irange>`.
4502cb42
SW
561 If the lower value specified happens to be larger than the upper value
562 the two values are swapped.
f80dba8d
MT
563
564.. _bool:
565
566**bool**
567 Boolean. Usually parsed as an integer, however only defined for
568 true and false (1 and 0).
569
570.. _irange:
571
572**irange**
573 Integer range with suffix. Allows value range to be given, such as
c60ebc45 574 1024-4096. A colon may also be used as the separator, e.g. 1k:4k. If the
f80dba8d
MT
575 option allows two sets of ranges, they can be specified with a ',' or '/'
576 delimiter: 1k-4k/8k-32k. Also see :ref:`int <int>`.
577
578**float_list**
579 A list of floating point numbers, separated by a ':' character.
580
f5c3bcf2
TK
581With the above in mind, here follows the complete list of fio job parameters.
582
f80dba8d
MT
583
584Units
585~~~~~
586
587.. option:: kb_base=int
588
589 Select the interpretation of unit prefixes in input parameters.
590
591 **1000**
592 Inputs comply with IEC 80000-13 and the International
593 System of Units (SI). Use:
594
595 - power-of-2 values with IEC prefixes (e.g., KiB)
596 - power-of-10 values with SI prefixes (e.g., kB)
597
598 **1024**
599 Compatibility mode (default). To avoid breaking old scripts:
600
601 - power-of-2 values with SI prefixes
602 - power-of-10 values with IEC prefixes
603
604 See :option:`bs` for more details on input parameters.
605
606 Outputs always use correct prefixes. Most outputs include both
607 side-by-side, like::
608
609 bw=2383.3kB/s (2327.4KiB/s)
610
611 If only one value is reported, then kb_base selects the one to use:
612
613 **1000** -- SI prefixes
614
615 **1024** -- IEC prefixes
616
617.. option:: unit_base=int
618
619 Base unit for reporting. Allowed values are:
620
621 **0**
622 Use auto-detection (default).
623 **8**
624 Byte based.
625 **1**
626 Bit based.
627
628
f80dba8d
MT
629Job description
630~~~~~~~~~~~~~~~
631
632.. option:: name=str
633
634 ASCII name of the job. This may be used to override the name printed by fio
635 for this job. Otherwise the job name is used. On the command line this
636 parameter has the special purpose of also signaling the start of a new job.
637
638.. option:: description=str
639
640 Text description of the job. Doesn't do anything except dump this text
641 description when this job is run. It's not parsed.
642
643.. option:: loops=int
644
645 Run the specified number of iterations of this job. Used to repeat the same
646 workload a given number of times. Defaults to 1.
647
648.. option:: numjobs=int
649
79591fa9
TK
650 Create the specified number of clones of this job. Each clone of job
651 is spawned as an independent thread or process. May be used to setup a
f80dba8d
MT
652 larger number of threads/processes doing the same thing. Each thread is
653 reported separately; to see statistics for all clones as a whole, use
654 :option:`group_reporting` in conjunction with :option:`new_group`.
a47b697c 655 See :option:`--max-jobs`. Default: 1.
f80dba8d
MT
656
657
658Time related parameters
659~~~~~~~~~~~~~~~~~~~~~~~
660
661.. option:: runtime=time
662
f75ede1d 663 Tell fio to terminate processing after the specified period of time. It
f80dba8d 664 can be quite hard to determine for how long a specified job will run, so
f75ede1d 665 this parameter is handy to cap the total runtime to a given time. When
947e0fe0 666 the unit is omitted, the value is intepreted in seconds.
f80dba8d
MT
667
668.. option:: time_based
669
670 If set, fio will run for the duration of the :option:`runtime` specified
671 even if the file(s) are completely read or written. It will simply loop over
672 the same workload as many times as the :option:`runtime` allows.
673
a881438b 674.. option:: startdelay=irange(time)
f80dba8d 675
947e0fe0
SW
676 Delay the start of job for the specified amount of time. Can be a single
677 value or a range. When given as a range, each thread will choose a value
678 randomly from within the range. Value is in seconds if a unit is omitted.
f80dba8d
MT
679
680.. option:: ramp_time=time
681
682 If set, fio will run the specified workload for this amount of time before
683 logging any performance numbers. Useful for letting performance settle
684 before logging results, thus minimizing the runtime required for stable
685 results. Note that the ``ramp_time`` is considered lead in time for a job,
686 thus it will increase the total runtime if a special timeout or
f75ede1d
SW
687 :option:`runtime` is specified. When the unit is omitted, the value is
688 given in seconds.
f80dba8d
MT
689
690.. option:: clocksource=str
691
692 Use the given clocksource as the base of timing. The supported options are:
693
694 **gettimeofday**
695 :manpage:`gettimeofday(2)`
696
697 **clock_gettime**
698 :manpage:`clock_gettime(2)`
699
700 **cpu**
701 Internal CPU clock source
702
703 cpu is the preferred clocksource if it is reliable, as it is very fast (and
704 fio is heavy on time calls). Fio will automatically use this clocksource if
705 it's supported and considered reliable on the system it is running on,
706 unless another clocksource is specifically set. For x86/x86-64 CPUs, this
707 means supporting TSC Invariant.
708
709.. option:: gtod_reduce=bool
710
711 Enable all of the :manpage:`gettimeofday(2)` reducing options
f75ede1d 712 (:option:`disable_clat`, :option:`disable_slat`, :option:`disable_bw_measurement`) plus
f80dba8d
MT
713 reduce precision of the timeout somewhat to really shrink the
714 :manpage:`gettimeofday(2)` call count. With this option enabled, we only do
715 about 0.4% of the :manpage:`gettimeofday(2)` calls we would have done if all
716 time keeping was enabled.
717
718.. option:: gtod_cpu=int
719
720 Sometimes it's cheaper to dedicate a single thread of execution to just
721 getting the current time. Fio (and databases, for instance) are very
722 intensive on :manpage:`gettimeofday(2)` calls. With this option, you can set
723 one CPU aside for doing nothing but logging current time to a shared memory
724 location. Then the other threads/processes that run I/O workloads need only
725 copy that segment, instead of entering the kernel with a
726 :manpage:`gettimeofday(2)` call. The CPU set aside for doing these time
727 calls will be excluded from other uses. Fio will manually clear it from the
728 CPU mask of other jobs.
729
730
731Target file/device
732~~~~~~~~~~~~~~~~~~
733
734.. option:: directory=str
735
736 Prefix filenames with this directory. Used to place files in a different
737 location than :file:`./`. You can specify a number of directories by
738 separating the names with a ':' character. These directories will be
02dd2689 739 assigned equally distributed to job clones created by :option:`numjobs` as
f80dba8d
MT
740 long as they are using generated filenames. If specific `filename(s)` are
741 set fio will use the first listed directory, and thereby matching the
742 `filename` semantic which generates a file each clone if not specified, but
743 let all clones use the same if set.
744
02dd2689
SW
745 See the :option:`filename` option for information on how to escape "``:``" and
746 "``\``" characters within the directory path itself.
f80dba8d
MT
747
748.. option:: filename=str
749
750 Fio normally makes up a `filename` based on the job name, thread number, and
02dd2689
SW
751 file number (see :option:`filename_format`). If you want to share files
752 between threads in a job or several
79591fa9
TK
753 jobs with fixed file paths, specify a `filename` for each of them to override
754 the default. If the ioengine is file based, you can specify a number of files
755 by separating the names with a ':' colon. So if you wanted a job to open
756 :file:`/dev/sda` and :file:`/dev/sdb` as the two working files, you would use
757 ``filename=/dev/sda:/dev/sdb``. This also means that whenever this option is
758 specified, :option:`nrfiles` is ignored. The size of regular files specified
02dd2689 759 by this option will be :option:`size` divided by number of files unless an
79591fa9
TK
760 explicit size is specified by :option:`filesize`.
761
02dd2689
SW
762 Each colon and backslash in the wanted path must be escaped with a ``\``
763 character. For instance, if the path is :file:`/dev/dsk/foo@3,0:c` then you
764 would use ``filename=/dev/dsk/foo@3,0\:c`` and if the path is
765 :file:`F:\\filename` then you would use ``filename=F\:\\filename``.
766
f80dba8d
MT
767 On Windows, disk devices are accessed as :file:`\\\\.\\PhysicalDrive0` for
768 the first device, :file:`\\\\.\\PhysicalDrive1` for the second etc.
769 Note: Windows and FreeBSD prevent write access to areas
02dd2689
SW
770 of the disk containing in-use data (e.g. filesystems).
771
772 The filename "`-`" is a reserved name, meaning *stdin* or *stdout*. Which
773 of the two depends on the read/write direction set.
f80dba8d
MT
774
775.. option:: filename_format=str
776
777 If sharing multiple files between jobs, it is usually necessary to have fio
778 generate the exact names that you want. By default, fio will name a file
779 based on the default file format specification of
780 :file:`jobname.jobnumber.filenumber`. With this option, that can be
781 customized. Fio will recognize and replace the following keywords in this
782 string:
783
784 **$jobname**
785 The name of the worker thread or process.
786 **$jobnum**
787 The incremental number of the worker thread or process.
788 **$filenum**
789 The incremental number of the file for that worker thread or
790 process.
791
792 To have dependent jobs share a set of files, this option can be set to have
793 fio generate filenames that are shared between the two. For instance, if
794 :file:`testfiles.$filenum` is specified, file number 4 for any job will be
795 named :file:`testfiles.4`. The default of :file:`$jobname.$jobnum.$filenum`
796 will be used if no other format specifier is given.
797
645943c0
JB
798 If you specify a path then the directories will be created up to the
799 main directory for the file. So for example if you specify
800 ``filename_format=a/b/c/$jobnum`` then the directories a/b/c will be
801 created before the file setup part of the job. If you specify
802 :option:`directory` then the path will be relative that directory,
803 otherwise it is treated as the absolute path.
804
f80dba8d
MT
805.. option:: unique_filename=bool
806
807 To avoid collisions between networked clients, fio defaults to prefixing any
808 generated filenames (with a directory specified) with the source of the
809 client connecting. To disable this behavior, set this option to 0.
810
811.. option:: opendir=str
812
813 Recursively open any files below directory `str`.
814
815.. option:: lockfile=str
816
817 Fio defaults to not locking any files before it does I/O to them. If a file
818 or file descriptor is shared, fio can serialize I/O to that file to make the
819 end result consistent. This is usual for emulating real workloads that share
820 files. The lock modes are:
821
822 **none**
823 No locking. The default.
824 **exclusive**
825 Only one thread or process may do I/O at a time, excluding all
826 others.
827 **readwrite**
828 Read-write locking on the file. Many readers may
829 access the file at the same time, but writes get exclusive access.
830
831.. option:: nrfiles=int
832
79591fa9
TK
833 Number of files to use for this job. Defaults to 1. The size of files
834 will be :option:`size` divided by this unless explicit size is specified by
835 :option:`filesize`. Files are created for each thread separately, and each
836 file will have a file number within its name by default, as explained in
837 :option:`filename` section.
838
f80dba8d
MT
839
840.. option:: openfiles=int
841
842 Number of files to keep open at the same time. Defaults to the same as
843 :option:`nrfiles`, can be set smaller to limit the number simultaneous
844 opens.
845
846.. option:: file_service_type=str
847
848 Defines how fio decides which file from a job to service next. The following
849 types are defined:
850
851 **random**
852 Choose a file at random.
853
854 **roundrobin**
855 Round robin over opened files. This is the default.
856
857 **sequential**
858 Finish one file before moving on to the next. Multiple files can
f50fbdda 859 still be open depending on :option:`openfiles`.
f80dba8d
MT
860
861 **zipf**
c60ebc45 862 Use a *Zipf* distribution to decide what file to access.
f80dba8d
MT
863
864 **pareto**
c60ebc45 865 Use a *Pareto* distribution to decide what file to access.
f80dba8d 866
dd3503d3 867 **normal**
c60ebc45 868 Use a *Gaussian* (normal) distribution to decide what file to
f80dba8d
MT
869 access.
870
dd3503d3
SW
871 **gauss**
872 Alias for normal.
873
f80dba8d
MT
874 For *random*, *roundrobin*, and *sequential*, a postfix can be appended to
875 tell fio how many I/Os to issue before switching to a new file. For example,
876 specifying ``file_service_type=random:8`` would cause fio to issue
877 8 I/Os before selecting a new file at random. For the non-uniform
878 distributions, a floating point postfix can be given to influence how the
879 distribution is skewed. See :option:`random_distribution` for a description
880 of how that would work.
881
882.. option:: ioscheduler=str
883
884 Attempt to switch the device hosting the file to the specified I/O scheduler
885 before running.
886
887.. option:: create_serialize=bool
888
889 If true, serialize the file creation for the jobs. This may be handy to
890 avoid interleaving of data files, which may greatly depend on the filesystem
a47b697c 891 used and even the number of processors in the system. Default: true.
f80dba8d
MT
892
893.. option:: create_fsync=bool
894
22413915 895 :manpage:`fsync(2)` the data file after creation. This is the default.
f80dba8d
MT
896
897.. option:: create_on_open=bool
898
730bd7d9
SW
899 If true, don't pre-create files but allow the job's open() to create a file
900 when it's time to do I/O. Default: false -- pre-create all necessary files
901 when the job starts.
f80dba8d
MT
902
903.. option:: create_only=bool
904
905 If true, fio will only run the setup phase of the job. If files need to be
4502cb42 906 laid out or updated on disk, only that will be done -- the actual job contents
a47b697c 907 are not executed. Default: false.
f80dba8d
MT
908
909.. option:: allow_file_create=bool
910
730bd7d9
SW
911 If true, fio is permitted to create files as part of its workload. If this
912 option is false, then fio will error out if
f80dba8d
MT
913 the files it needs to use don't already exist. Default: true.
914
915.. option:: allow_mounted_write=bool
916
c60ebc45 917 If this isn't set, fio will abort jobs that are destructive (e.g. that write)
f80dba8d
MT
918 to what appears to be a mounted device or partition. This should help catch
919 creating inadvertently destructive tests, not realizing that the test will
b1db0375
TK
920 destroy data on the mounted file system. Note that some platforms don't allow
921 writing against a mounted device regardless of this option. Default: false.
f80dba8d
MT
922
923.. option:: pre_read=bool
924
925 If this is given, files will be pre-read into memory before starting the
926 given I/O operation. This will also clear the :option:`invalidate` flag,
927 since it is pointless to pre-read and then drop the cache. This will only
928 work for I/O engines that are seek-able, since they allow you to read the
a47b697c
SW
929 same data multiple times. Thus it will not work on non-seekable I/O engines
930 (e.g. network, splice). Default: false.
f80dba8d
MT
931
932.. option:: unlink=bool
933
934 Unlink the job files when done. Not the default, as repeated runs of that
a47b697c
SW
935 job would then waste time recreating the file set again and again. Default:
936 false.
f80dba8d
MT
937
938.. option:: unlink_each_loop=bool
939
a47b697c 940 Unlink job files after each iteration or loop. Default: false.
f80dba8d
MT
941
942.. option:: zonesize=int
943
944 Divide a file into zones of the specified size. See :option:`zoneskip`.
945
946.. option:: zonerange=int
947
948 Give size of an I/O zone. See :option:`zoneskip`.
949
950.. option:: zoneskip=int
951
952 Skip the specified number of bytes when :option:`zonesize` data has been
953 read. The two zone options can be used to only do I/O on zones of a file.
954
955
956I/O type
957~~~~~~~~
958
959.. option:: direct=bool
960
961 If value is true, use non-buffered I/O. This is usually O_DIRECT. Note that
8e889110 962 OpenBSD and ZFS on Solaris don't support direct I/O. On Windows the synchronous
f80dba8d
MT
963 ioengines don't support direct I/O. Default: false.
964
965.. option:: atomic=bool
966
967 If value is true, attempt to use atomic direct I/O. Atomic writes are
968 guaranteed to be stable once acknowledged by the operating system. Only
969 Linux supports O_ATOMIC right now.
970
971.. option:: buffered=bool
972
973 If value is true, use buffered I/O. This is the opposite of the
974 :option:`direct` option. Defaults to true.
975
976.. option:: readwrite=str, rw=str
977
978 Type of I/O pattern. Accepted values are:
979
980 **read**
981 Sequential reads.
982 **write**
983 Sequential writes.
984 **trim**
985 Sequential trims (Linux block devices only).
f80dba8d
MT
986 **randread**
987 Random reads.
2831be97
SW
988 **randwrite**
989 Random writes.
f80dba8d
MT
990 **randtrim**
991 Random trims (Linux block devices only).
992 **rw,readwrite**
993 Sequential mixed reads and writes.
994 **randrw**
995 Random mixed reads and writes.
996 **trimwrite**
997 Sequential trim+write sequences. Blocks will be trimmed first,
998 then the same blocks will be written to.
999
1000 Fio defaults to read if the option is not specified. For the mixed I/O
1001 types, the default is to split them 50/50. For certain types of I/O the
730bd7d9
SW
1002 result may still be skewed a bit, since the speed may be different.
1003
1004 It is possible to specify the number of I/Os to do before getting a new
1005 offset by appending ``:<nr>`` to the end of the string given. For a
f80dba8d
MT
1006 random read, it would look like ``rw=randread:8`` for passing in an offset
1007 modifier with a value of 8. If the suffix is used with a sequential I/O
730bd7d9
SW
1008 pattern, then the *<nr>* value specified will be **added** to the generated
1009 offset for each I/O turning sequential I/O into sequential I/O with holes.
1010 For instance, using ``rw=write:4k`` will skip 4k for every write. Also see
1011 the :option:`rw_sequencer` option.
f80dba8d
MT
1012
1013.. option:: rw_sequencer=str
1014
1015 If an offset modifier is given by appending a number to the ``rw=<str>``
1016 line, then this option controls how that number modifies the I/O offset
1017 being generated. Accepted values are:
1018
1019 **sequential**
1020 Generate sequential offset.
1021 **identical**
1022 Generate the same offset.
1023
1024 ``sequential`` is only useful for random I/O, where fio would normally
c60ebc45 1025 generate a new random offset for every I/O. If you append e.g. 8 to randread,
007c7be9
SW
1026 you would get a new random offset for every 8 I/Os. The result would be a
1027 seek for only every 8 I/Os, instead of for every I/O. Use ``rw=randread:8``
f80dba8d
MT
1028 to specify that. As sequential I/O is already sequential, setting
1029 ``sequential`` for that would not result in any differences. ``identical``
1030 behaves in a similar fashion, except it sends the same offset 8 number of
1031 times before generating a new offset.
1032
1033.. option:: unified_rw_reporting=bool
1034
1035 Fio normally reports statistics on a per data direction basis, meaning that
1036 reads, writes, and trims are accounted and reported separately. If this
1037 option is set fio sums the results and report them as "mixed" instead.
1038
1039.. option:: randrepeat=bool
1040
1041 Seed the random number generator used for random I/O patterns in a
1042 predictable way so the pattern is repeatable across runs. Default: true.
1043
1044.. option:: allrandrepeat=bool
1045
1046 Seed all random number generators in a predictable way so results are
1047 repeatable across runs. Default: false.
1048
1049.. option:: randseed=int
1050
1051 Seed the random number generators based on this seed value, to be able to
1052 control what sequence of output is being generated. If not set, the random
1053 sequence depends on the :option:`randrepeat` setting.
1054
1055.. option:: fallocate=str
1056
1057 Whether pre-allocation is performed when laying down files.
1058 Accepted values are:
1059
1060 **none**
1061 Do not pre-allocate space.
1062
2c3e17be
SW
1063 **native**
1064 Use a platform's native pre-allocation call but fall back to
1065 **none** behavior if it fails/is not implemented.
1066
f80dba8d
MT
1067 **posix**
1068 Pre-allocate via :manpage:`posix_fallocate(3)`.
1069
1070 **keep**
1071 Pre-allocate via :manpage:`fallocate(2)` with
1072 FALLOC_FL_KEEP_SIZE set.
1073
1074 **0**
1075 Backward-compatible alias for **none**.
1076
1077 **1**
1078 Backward-compatible alias for **posix**.
1079
1080 May not be available on all supported platforms. **keep** is only available
2c3e17be
SW
1081 on Linux. If using ZFS on Solaris this cannot be set to **posix**
1082 because ZFS doesn't support pre-allocation. Default: **native** if any
1083 pre-allocation methods are available, **none** if not.
f80dba8d
MT
1084
1085.. option:: fadvise_hint=str
1086
1087 Use :manpage:`posix_fadvise(2)` to advise the kernel on what I/O patterns
1088 are likely to be issued. Accepted values are:
1089
1090 **0**
1091 Backwards-compatible hint for "no hint".
1092
1093 **1**
1094 Backwards compatible hint for "advise with fio workload type". This
1095 uses **FADV_RANDOM** for a random workload, and **FADV_SEQUENTIAL**
1096 for a sequential workload.
1097
1098 **sequential**
1099 Advise using **FADV_SEQUENTIAL**.
1100
1101 **random**
1102 Advise using **FADV_RANDOM**.
1103
8f4b9f24 1104.. option:: write_hint=str
f80dba8d 1105
8f4b9f24
JA
1106 Use :manpage:`fcntl(2)` to advise the kernel what life time to expect
1107 from a write. Only supported on Linux, as of version 4.13. Accepted
1108 values are:
1109
1110 **none**
1111 No particular life time associated with this file.
1112
1113 **short**
1114 Data written to this file has a short life time.
1115
1116 **medium**
1117 Data written to this file has a medium life time.
1118
1119 **long**
1120 Data written to this file has a long life time.
1121
1122 **extreme**
1123 Data written to this file has a very long life time.
1124
1125 The values are all relative to each other, and no absolute meaning
1126 should be associated with them.
f80dba8d
MT
1127
1128.. option:: offset=int
1129
82dbb8cb 1130 Start I/O at the provided offset in the file, given as either a fixed size in
83c8b093
JF
1131 bytes or a percentage. If a percentage is given, the generated offset will be
1132 aligned to the minimum ``blocksize`` or to the value of ``offset_align`` if
1133 provided. Data before the given offset will not be touched. This
89978a6b
BW
1134 effectively caps the file size at `real_size - offset`. Can be combined with
1135 :option:`size` to constrain the start and end range of the I/O workload.
44bb1142
TK
1136 A percentage can be specified by a number between 1 and 100 followed by '%',
1137 for example, ``offset=20%`` to specify 20%.
f80dba8d 1138
83c8b093
JF
1139.. option:: offset_align=int
1140
1141 If set to non-zero value, the byte offset generated by a percentage ``offset``
1142 is aligned upwards to this value. Defaults to 0 meaning that a percentage
1143 offset is aligned to the minimum block size.
1144
f80dba8d
MT
1145.. option:: offset_increment=int
1146
1147 If this is provided, then the real offset becomes `offset + offset_increment
1148 * thread_number`, where the thread number is a counter that starts at 0 and
1149 is incremented for each sub-job (i.e. when :option:`numjobs` option is
1150 specified). This option is useful if there are several jobs which are
1151 intended to operate on a file in parallel disjoint segments, with even
1152 spacing between the starting points.
1153
1154.. option:: number_ios=int
1155
c60ebc45 1156 Fio will normally perform I/Os until it has exhausted the size of the region
f80dba8d
MT
1157 set by :option:`size`, or if it exhaust the allocated time (or hits an error
1158 condition). With this setting, the range/size can be set independently of
c60ebc45 1159 the number of I/Os to perform. When fio reaches this number, it will exit
f80dba8d
MT
1160 normally and report status. Note that this does not extend the amount of I/O
1161 that will be done, it will only stop fio if this condition is met before
1162 other end-of-job criteria.
1163
1164.. option:: fsync=int
1165
730bd7d9
SW
1166 If writing to a file, issue an :manpage:`fsync(2)` (or its equivalent) of
1167 the dirty data for every number of blocks given. For example, if you give 32
1168 as a parameter, fio will sync the file after every 32 writes issued. If fio is
1169 using non-buffered I/O, we may not sync the file. The exception is the sg
1170 I/O engine, which synchronizes the disk cache anyway. Defaults to 0, which
1171 means fio does not periodically issue and wait for a sync to complete. Also
1172 see :option:`end_fsync` and :option:`fsync_on_close`.
f80dba8d
MT
1173
1174.. option:: fdatasync=int
1175
1176 Like :option:`fsync` but uses :manpage:`fdatasync(2)` to only sync data and
000a5f1c 1177 not metadata blocks. In Windows, FreeBSD, and DragonFlyBSD there is no
730bd7d9
SW
1178 :manpage:`fdatasync(2)` so this falls back to using :manpage:`fsync(2)`.
1179 Defaults to 0, which means fio does not periodically issue and wait for a
1180 data-only sync to complete.
f80dba8d
MT
1181
1182.. option:: write_barrier=int
1183
2831be97 1184 Make every `N-th` write a barrier write.
f80dba8d 1185
f50fbdda 1186.. option:: sync_file_range=str:int
f80dba8d 1187
f50fbdda 1188 Use :manpage:`sync_file_range(2)` for every `int` number of write
f80dba8d
MT
1189 operations. Fio will track range of writes that have happened since the last
1190 :manpage:`sync_file_range(2)` call. `str` can currently be one or more of:
1191
1192 **wait_before**
1193 SYNC_FILE_RANGE_WAIT_BEFORE
1194 **write**
1195 SYNC_FILE_RANGE_WRITE
1196 **wait_after**
1197 SYNC_FILE_RANGE_WAIT_AFTER
1198
1199 So if you do ``sync_file_range=wait_before,write:8``, fio would use
1200 ``SYNC_FILE_RANGE_WAIT_BEFORE | SYNC_FILE_RANGE_WRITE`` for every 8
1201 writes. Also see the :manpage:`sync_file_range(2)` man page. This option is
1202 Linux specific.
1203
1204.. option:: overwrite=bool
1205
1206 If true, writes to a file will always overwrite existing data. If the file
1207 doesn't already exist, it will be created before the write phase begins. If
1208 the file exists and is large enough for the specified write phase, nothing
a47b697c 1209 will be done. Default: false.
f80dba8d
MT
1210
1211.. option:: end_fsync=bool
1212
a47b697c
SW
1213 If true, :manpage:`fsync(2)` file contents when a write stage has completed.
1214 Default: false.
f80dba8d
MT
1215
1216.. option:: fsync_on_close=bool
1217
1218 If true, fio will :manpage:`fsync(2)` a dirty file on close. This differs
a47b697c
SW
1219 from :option:`end_fsync` in that it will happen on every file close, not
1220 just at the end of the job. Default: false.
f80dba8d
MT
1221
1222.. option:: rwmixread=int
1223
1224 Percentage of a mixed workload that should be reads. Default: 50.
1225
1226.. option:: rwmixwrite=int
1227
1228 Percentage of a mixed workload that should be writes. If both
1229 :option:`rwmixread` and :option:`rwmixwrite` is given and the values do not
1230 add up to 100%, the latter of the two will be used to override the
1231 first. This may interfere with a given rate setting, if fio is asked to
1232 limit reads or writes to a certain rate. If that is the case, then the
1233 distribution may be skewed. Default: 50.
1234
1235.. option:: random_distribution=str:float[,str:float][,str:float]
1236
1237 By default, fio will use a completely uniform random distribution when asked
1238 to perform random I/O. Sometimes it is useful to skew the distribution in
1239 specific ways, ensuring that some parts of the data is more hot than others.
1240 fio includes the following distribution models:
1241
1242 **random**
1243 Uniform random distribution
1244
1245 **zipf**
1246 Zipf distribution
1247
1248 **pareto**
1249 Pareto distribution
1250
b2f4b559 1251 **normal**
c60ebc45 1252 Normal (Gaussian) distribution
f80dba8d
MT
1253
1254 **zoned**
1255 Zoned random distribution
1256
59466396
JA
1257 **zoned_abs**
1258 Zone absolute random distribution
1259
f80dba8d 1260 When using a **zipf** or **pareto** distribution, an input value is also
f50fbdda 1261 needed to define the access pattern. For **zipf**, this is the `Zipf
c60ebc45 1262 theta`. For **pareto**, it's the `Pareto power`. Fio includes a test
f50fbdda 1263 program, :command:`fio-genzipf`, that can be used visualize what the given input
f80dba8d
MT
1264 values will yield in terms of hit rates. If you wanted to use **zipf** with
1265 a `theta` of 1.2, you would use ``random_distribution=zipf:1.2`` as the
1266 option. If a non-uniform model is used, fio will disable use of the random
b2f4b559
SW
1267 map. For the **normal** distribution, a normal (Gaussian) deviation is
1268 supplied as a value between 0 and 100.
f80dba8d
MT
1269
1270 For a **zoned** distribution, fio supports specifying percentages of I/O
1271 access that should fall within what range of the file or device. For
1272 example, given a criteria of:
1273
f50fbdda
TK
1274 * 60% of accesses should be to the first 10%
1275 * 30% of accesses should be to the next 20%
1276 * 8% of accesses should be to the next 30%
1277 * 2% of accesses should be to the next 40%
f80dba8d
MT
1278
1279 we can define that through zoning of the random accesses. For the above
1280 example, the user would do::
1281
1282 random_distribution=zoned:60/10:30/20:8/30:2/40
1283
59466396
JA
1284 A **zoned_abs** distribution works exactly like the **zoned**, except
1285 that it takes absolute sizes. For example, let's say you wanted to
1286 define access according to the following criteria:
1287
1288 * 60% of accesses should be to the first 20G
1289 * 30% of accesses should be to the next 100G
1290 * 10% of accesses should be to the next 500G
1291
1292 we can define an absolute zoning distribution with:
1293
1294 random_distribution=zoned_abs=60/20G:30/100G:10/500g
1295
1296 Similarly to how :option:`bssplit` works for setting ranges and
1297 percentages of block sizes. Like :option:`bssplit`, it's possible to
1298 specify separate zones for reads, writes, and trims. If just one set
1299 is given, it'll apply to all of them. This goes for both **zoned**
1300 **zoned_abs** distributions.
f80dba8d
MT
1301
1302.. option:: percentage_random=int[,int][,int]
1303
1304 For a random workload, set how big a percentage should be random. This
1305 defaults to 100%, in which case the workload is fully random. It can be set
1306 from anywhere from 0 to 100. Setting it to 0 would make the workload fully
1307 sequential. Any setting in between will result in a random mix of sequential
1308 and random I/O, at the given percentages. Comma-separated values may be
1309 specified for reads, writes, and trims as described in :option:`blocksize`.
1310
1311.. option:: norandommap
1312
1313 Normally fio will cover every block of the file when doing random I/O. If
1314 this option is given, fio will just get a new random offset without looking
1315 at past I/O history. This means that some blocks may not be read or written,
1316 and that some blocks may be read/written more than once. If this option is
1317 used with :option:`verify` and multiple blocksizes (via :option:`bsrange`),
1318 only intact blocks are verified, i.e., partially-overwritten blocks are
1319 ignored.
1320
1321.. option:: softrandommap=bool
1322
1323 See :option:`norandommap`. If fio runs with the random block map enabled and
1324 it fails to allocate the map, if this option is set it will continue without
1325 a random block map. As coverage will not be as complete as with random maps,
1326 this option is disabled by default.
1327
1328.. option:: random_generator=str
1329
f50fbdda 1330 Fio supports the following engines for generating I/O offsets for random I/O:
f80dba8d
MT
1331
1332 **tausworthe**
f50fbdda 1333 Strong 2^88 cycle random number generator.
f80dba8d 1334 **lfsr**
f50fbdda 1335 Linear feedback shift register generator.
f80dba8d 1336 **tausworthe64**
f50fbdda 1337 Strong 64-bit 2^258 cycle random number generator.
f80dba8d
MT
1338
1339 **tausworthe** is a strong random number generator, but it requires tracking
1340 on the side if we want to ensure that blocks are only read or written
f50fbdda 1341 once. **lfsr** guarantees that we never generate the same offset twice, and
f80dba8d 1342 it's also less computationally expensive. It's not a true random generator,
f50fbdda 1343 however, though for I/O purposes it's typically good enough. **lfsr** only
f80dba8d
MT
1344 works with single block sizes, not with workloads that use multiple block
1345 sizes. If used with such a workload, fio may read or write some blocks
1346 multiple times. The default value is **tausworthe**, unless the required
1347 space exceeds 2^32 blocks. If it does, then **tausworthe64** is
1348 selected automatically.
1349
1350
1351Block size
1352~~~~~~~~~~
1353
1354.. option:: blocksize=int[,int][,int], bs=int[,int][,int]
1355
1356 The block size in bytes used for I/O units. Default: 4096. A single value
1357 applies to reads, writes, and trims. Comma-separated values may be
1358 specified for reads, writes, and trims. A value not terminated in a comma
1359 applies to subsequent types.
1360
1361 Examples:
1362
1363 **bs=256k**
1364 means 256k for reads, writes and trims.
1365
1366 **bs=8k,32k**
1367 means 8k for reads, 32k for writes and trims.
1368
1369 **bs=8k,32k,**
1370 means 8k for reads, 32k for writes, and default for trims.
1371
1372 **bs=,8k**
1373 means default for reads, 8k for writes and trims.
1374
1375 **bs=,8k,**
b443ae44 1376 means default for reads, 8k for writes, and default for trims.
f80dba8d
MT
1377
1378.. option:: blocksize_range=irange[,irange][,irange], bsrange=irange[,irange][,irange]
1379
1380 A range of block sizes in bytes for I/O units. The issued I/O unit will
1381 always be a multiple of the minimum size, unless
1382 :option:`blocksize_unaligned` is set.
1383
1384 Comma-separated ranges may be specified for reads, writes, and trims as
1385 described in :option:`blocksize`.
1386
1387 Example: ``bsrange=1k-4k,2k-8k``.
1388
1389.. option:: bssplit=str[,str][,str]
1390
1391 Sometimes you want even finer grained control of the block sizes issued, not
1392 just an even split between them. This option allows you to weight various
1393 block sizes, so that you are able to define a specific amount of block sizes
1394 issued. The format for this option is::
1395
1396 bssplit=blocksize/percentage:blocksize/percentage
1397
1398 for as many block sizes as needed. So if you want to define a workload that
1399 has 50% 64k blocks, 10% 4k blocks, and 40% 32k blocks, you would write::
1400
1401 bssplit=4k/10:64k/50:32k/40
1402
1403 Ordering does not matter. If the percentage is left blank, fio will fill in
1404 the remaining values evenly. So a bssplit option like this one::
1405
1406 bssplit=4k/50:1k/:32k/
1407
1408 would have 50% 4k ios, and 25% 1k and 32k ios. The percentages always add up
1409 to 100, if bssplit is given a range that adds up to more, it will error out.
1410
1411 Comma-separated values may be specified for reads, writes, and trims as
1412 described in :option:`blocksize`.
1413
1414 If you want a workload that has 50% 2k reads and 50% 4k reads, while having
1415 90% 4k writes and 10% 8k writes, you would specify::
1416
1417 bssplit=2k/50:4k/50,4k/90,8k/10
1418
1419.. option:: blocksize_unaligned, bs_unaligned
1420
1421 If set, fio will issue I/O units with any size within
1422 :option:`blocksize_range`, not just multiples of the minimum size. This
1423 typically won't work with direct I/O, as that normally requires sector
1424 alignment.
1425
589e88b7 1426.. option:: bs_is_seq_rand=bool
f80dba8d
MT
1427
1428 If this option is set, fio will use the normal read,write blocksize settings
1429 as sequential,random blocksize settings instead. Any random read or write
1430 will use the WRITE blocksize settings, and any sequential read or write will
1431 use the READ blocksize settings.
1432
1433.. option:: blockalign=int[,int][,int], ba=int[,int][,int]
1434
1435 Boundary to which fio will align random I/O units. Default:
1436 :option:`blocksize`. Minimum alignment is typically 512b for using direct
1437 I/O, though it usually depends on the hardware block size. This option is
1438 mutually exclusive with using a random map for files, so it will turn off
1439 that option. Comma-separated values may be specified for reads, writes, and
1440 trims as described in :option:`blocksize`.
1441
1442
1443Buffers and memory
1444~~~~~~~~~~~~~~~~~~
1445
1446.. option:: zero_buffers
1447
1448 Initialize buffers with all zeros. Default: fill buffers with random data.
1449
1450.. option:: refill_buffers
1451
1452 If this option is given, fio will refill the I/O buffers on every
72592780
SW
1453 submit. Only makes sense if :option:`zero_buffers` isn't specified,
1454 naturally. Defaults to being unset i.e., the buffer is only filled at
1455 init time and the data in it is reused when possible but if any of
1456 :option:`verify`, :option:`buffer_compress_percentage` or
1457 :option:`dedupe_percentage` are enabled then `refill_buffers` is also
1458 automatically enabled.
f80dba8d
MT
1459
1460.. option:: scramble_buffers=bool
1461
1462 If :option:`refill_buffers` is too costly and the target is using data
1463 deduplication, then setting this option will slightly modify the I/O buffer
1464 contents to defeat normal de-dupe attempts. This is not enough to defeat
1465 more clever block compression attempts, but it will stop naive dedupe of
1466 blocks. Default: true.
1467
1468.. option:: buffer_compress_percentage=int
1469
72592780
SW
1470 If this is set, then fio will attempt to provide I/O buffer content
1471 (on WRITEs) that compresses to the specified level. Fio does this by
1472 providing a mix of random data followed by fixed pattern data. The
1473 fixed pattern is either zeros, or the pattern specified by
1474 :option:`buffer_pattern`. If the `buffer_pattern` option is used, it
1475 might skew the compression ratio slightly. Setting
1476 `buffer_compress_percentage` to a value other than 100 will also
1477 enable :option:`refill_buffers` in order to reduce the likelihood that
1478 adjacent blocks are so similar that they over compress when seen
1479 together. See :option:`buffer_compress_chunk` for how to set a finer or
1480 coarser granularity for the random/fixed data region. Defaults to unset
1481 i.e., buffer data will not adhere to any compression level.
f80dba8d
MT
1482
1483.. option:: buffer_compress_chunk=int
1484
72592780
SW
1485 This setting allows fio to manage how big the random/fixed data region
1486 is when using :option:`buffer_compress_percentage`. When
1487 `buffer_compress_chunk` is set to some non-zero value smaller than the
1488 block size, fio can repeat the random/fixed region throughout the I/O
1489 buffer at the specified interval (which particularly useful when
1490 bigger block sizes are used for a job). When set to 0, fio will use a
1491 chunk size that matches the block size resulting in a single
1492 random/fixed region within the I/O buffer. Defaults to 512. When the
1493 unit is omitted, the value is interpreted in bytes.
f80dba8d
MT
1494
1495.. option:: buffer_pattern=str
1496
a1554f65
SB
1497 If set, fio will fill the I/O buffers with this pattern or with the contents
1498 of a file. If not set, the contents of I/O buffers are defined by the other
1499 options related to buffer contents. The setting can be any pattern of bytes,
1500 and can be prefixed with 0x for hex values. It may also be a string, where
1501 the string must then be wrapped with ``""``. Or it may also be a filename,
1502 where the filename must be wrapped with ``''`` in which case the file is
1503 opened and read. Note that not all the file contents will be read if that
1504 would cause the buffers to overflow. So, for example::
1505
1506 buffer_pattern='filename'
1507
1508 or::
f80dba8d
MT
1509
1510 buffer_pattern="abcd"
1511
1512 or::
1513
1514 buffer_pattern=-12
1515
1516 or::
1517
1518 buffer_pattern=0xdeadface
1519
1520 Also you can combine everything together in any order::
1521
a1554f65 1522 buffer_pattern=0xdeadface"abcd"-12'filename'
f80dba8d
MT
1523
1524.. option:: dedupe_percentage=int
1525
1526 If set, fio will generate this percentage of identical buffers when
1527 writing. These buffers will be naturally dedupable. The contents of the
1528 buffers depend on what other buffer compression settings have been set. It's
1529 possible to have the individual buffers either fully compressible, or not at
72592780
SW
1530 all -- this option only controls the distribution of unique buffers. Setting
1531 this option will also enable :option:`refill_buffers` to prevent every buffer
1532 being identical.
f80dba8d
MT
1533
1534.. option:: invalidate=bool
1535
730bd7d9
SW
1536 Invalidate the buffer/page cache parts of the files to be used prior to
1537 starting I/O if the platform and file type support it. Defaults to true.
21c1b29e
TK
1538 This will be ignored if :option:`pre_read` is also specified for the
1539 same job.
f80dba8d
MT
1540
1541.. option:: sync=bool
1542
1543 Use synchronous I/O for buffered writes. For the majority of I/O engines,
1544 this means using O_SYNC. Default: false.
1545
1546.. option:: iomem=str, mem=str
1547
1548 Fio can use various types of memory as the I/O unit buffer. The allowed
1549 values are:
1550
1551 **malloc**
1552 Use memory from :manpage:`malloc(3)` as the buffers. Default memory
1553 type.
1554
1555 **shm**
1556 Use shared memory as the buffers. Allocated through
1557 :manpage:`shmget(2)`.
1558
1559 **shmhuge**
1560 Same as shm, but use huge pages as backing.
1561
1562 **mmap**
22413915 1563 Use :manpage:`mmap(2)` to allocate buffers. May either be anonymous memory, or can
f80dba8d
MT
1564 be file backed if a filename is given after the option. The format
1565 is `mem=mmap:/path/to/file`.
1566
1567 **mmaphuge**
1568 Use a memory mapped huge file as the buffer backing. Append filename
1569 after mmaphuge, ala `mem=mmaphuge:/hugetlbfs/file`.
1570
1571 **mmapshared**
1572 Same as mmap, but use a MMAP_SHARED mapping.
1573
03553853
YR
1574 **cudamalloc**
1575 Use GPU memory as the buffers for GPUDirect RDMA benchmark.
f50fbdda 1576 The :option:`ioengine` must be `rdma`.
03553853 1577
f80dba8d
MT
1578 The area allocated is a function of the maximum allowed bs size for the job,
1579 multiplied by the I/O depth given. Note that for **shmhuge** and
1580 **mmaphuge** to work, the system must have free huge pages allocated. This
1581 can normally be checked and set by reading/writing
1582 :file:`/proc/sys/vm/nr_hugepages` on a Linux system. Fio assumes a huge page
1583 is 4MiB in size. So to calculate the number of huge pages you need for a
1584 given job file, add up the I/O depth of all jobs (normally one unless
1585 :option:`iodepth` is used) and multiply by the maximum bs set. Then divide
1586 that number by the huge page size. You can see the size of the huge pages in
1587 :file:`/proc/meminfo`. If no huge pages are allocated by having a non-zero
1588 number in `nr_hugepages`, using **mmaphuge** or **shmhuge** will fail. Also
1589 see :option:`hugepage-size`.
1590
1591 **mmaphuge** also needs to have hugetlbfs mounted and the file location
1592 should point there. So if it's mounted in :file:`/huge`, you would use
1593 `mem=mmaphuge:/huge/somefile`.
1594
f50fbdda 1595.. option:: iomem_align=int, mem_align=int
f80dba8d
MT
1596
1597 This indicates the memory alignment of the I/O memory buffers. Note that
1598 the given alignment is applied to the first I/O unit buffer, if using
1599 :option:`iodepth` the alignment of the following buffers are given by the
1600 :option:`bs` used. In other words, if using a :option:`bs` that is a
1601 multiple of the page sized in the system, all buffers will be aligned to
1602 this value. If using a :option:`bs` that is not page aligned, the alignment
1603 of subsequent I/O memory buffers is the sum of the :option:`iomem_align` and
1604 :option:`bs` used.
1605
1606.. option:: hugepage-size=int
1607
1608 Defines the size of a huge page. Must at least be equal to the system
1609 setting, see :file:`/proc/meminfo`. Defaults to 4MiB. Should probably
1610 always be a multiple of megabytes, so using ``hugepage-size=Xm`` is the
1611 preferred way to set this to avoid setting a non-pow-2 bad value.
1612
1613.. option:: lockmem=int
1614
1615 Pin the specified amount of memory with :manpage:`mlock(2)`. Can be used to
1616 simulate a smaller amount of memory. The amount specified is per worker.
1617
1618
1619I/O size
1620~~~~~~~~
1621
1622.. option:: size=int
1623
79591fa9
TK
1624 The total size of file I/O for each thread of this job. Fio will run until
1625 this many bytes has been transferred, unless runtime is limited by other options
1626 (such as :option:`runtime`, for instance, or increased/decreased by :option:`io_size`).
1627 Fio will divide this size between the available files determined by options
1628 such as :option:`nrfiles`, :option:`filename`, unless :option:`filesize` is
1629 specified by the job. If the result of division happens to be 0, the size is
c4aa2d08 1630 set to the physical size of the given files or devices if they exist.
79591fa9 1631 If this option is not specified, fio will use the full size of the given
f80dba8d
MT
1632 files or devices. If the files do not exist, size must be given. It is also
1633 possible to give size as a percentage between 1 and 100. If ``size=20%`` is
1634 given, fio will use 20% of the full size of the given files or devices.
9d25d068
SW
1635 Can be combined with :option:`offset` to constrain the start and end range
1636 that I/O will be done within.
f80dba8d
MT
1637
1638.. option:: io_size=int, io_limit=int
1639
1640 Normally fio operates within the region set by :option:`size`, which means
1641 that the :option:`size` option sets both the region and size of I/O to be
1642 performed. Sometimes that is not what you want. With this option, it is
1643 possible to define just the amount of I/O that fio should do. For instance,
1644 if :option:`size` is set to 20GiB and :option:`io_size` is set to 5GiB, fio
1645 will perform I/O within the first 20GiB but exit when 5GiB have been
1646 done. The opposite is also possible -- if :option:`size` is set to 20GiB,
1647 and :option:`io_size` is set to 40GiB, then fio will do 40GiB of I/O within
1648 the 0..20GiB region.
1649
7fdd97ca 1650.. option:: filesize=irange(int)
f80dba8d
MT
1651
1652 Individual file sizes. May be a range, in which case fio will select sizes
1653 for files at random within the given range and limited to :option:`size` in
1654 total (if that is given). If not given, each created file is the same size.
79591fa9
TK
1655 This option overrides :option:`size` in terms of file size, which means
1656 this value is used as a fixed size or possible range of each file.
f80dba8d
MT
1657
1658.. option:: file_append=bool
1659
1660 Perform I/O after the end of the file. Normally fio will operate within the
1661 size of a file. If this option is set, then fio will append to the file
1662 instead. This has identical behavior to setting :option:`offset` to the size
1663 of a file. This option is ignored on non-regular files.
1664
1665.. option:: fill_device=bool, fill_fs=bool
1666
1667 Sets size to something really large and waits for ENOSPC (no space left on
1668 device) as the terminating condition. Only makes sense with sequential
1669 write. For a read workload, the mount point will be filled first then I/O
1670 started on the result. This option doesn't make sense if operating on a raw
1671 device node, since the size of that is already known by the file system.
1672 Additionally, writing beyond end-of-device will not return ENOSPC there.
1673
1674
1675I/O engine
1676~~~~~~~~~~
1677
1678.. option:: ioengine=str
1679
1680 Defines how the job issues I/O to the file. The following types are defined:
1681
1682 **sync**
1683 Basic :manpage:`read(2)` or :manpage:`write(2)`
1684 I/O. :manpage:`lseek(2)` is used to position the I/O location.
54227e6b 1685 See :option:`fsync` and :option:`fdatasync` for syncing write I/Os.
f80dba8d
MT
1686
1687 **psync**
1688 Basic :manpage:`pread(2)` or :manpage:`pwrite(2)` I/O. Default on
1689 all supported operating systems except for Windows.
1690
1691 **vsync**
1692 Basic :manpage:`readv(2)` or :manpage:`writev(2)` I/O. Will emulate
c60ebc45 1693 queuing by coalescing adjacent I/Os into a single submission.
f80dba8d
MT
1694
1695 **pvsync**
1696 Basic :manpage:`preadv(2)` or :manpage:`pwritev(2)` I/O.
1697
1698 **pvsync2**
1699 Basic :manpage:`preadv2(2)` or :manpage:`pwritev2(2)` I/O.
1700
1701 **libaio**
1702 Linux native asynchronous I/O. Note that Linux may only support
22413915 1703 queued behavior with non-buffered I/O (set ``direct=1`` or
f80dba8d
MT
1704 ``buffered=0``).
1705 This engine defines engine specific options.
1706
1707 **posixaio**
1708 POSIX asynchronous I/O using :manpage:`aio_read(3)` and
1709 :manpage:`aio_write(3)`.
1710
1711 **solarisaio**
1712 Solaris native asynchronous I/O.
1713
1714 **windowsaio**
1715 Windows native asynchronous I/O. Default on Windows.
1716
1717 **mmap**
1718 File is memory mapped with :manpage:`mmap(2)` and data copied
1719 to/from using :manpage:`memcpy(3)`.
1720
1721 **splice**
1722 :manpage:`splice(2)` is used to transfer the data and
1723 :manpage:`vmsplice(2)` to transfer data from user space to the
1724 kernel.
1725
1726 **sg**
1727 SCSI generic sg v3 I/O. May either be synchronous using the SG_IO
1728 ioctl, or if the target is an sg character device we use
1729 :manpage:`read(2)` and :manpage:`write(2)` for asynchronous
f50fbdda
TK
1730 I/O. Requires :option:`filename` option to specify either block or
1731 character devices.
f80dba8d
MT
1732
1733 **null**
1734 Doesn't transfer any data, just pretends to. This is mainly used to
1735 exercise fio itself and for debugging/testing purposes.
1736
1737 **net**
1738 Transfer over the network to given ``host:port``. Depending on the
1739 :option:`protocol` used, the :option:`hostname`, :option:`port`,
1740 :option:`listen` and :option:`filename` options are used to specify
1741 what sort of connection to make, while the :option:`protocol` option
1742 determines which protocol will be used. This engine defines engine
1743 specific options.
1744
1745 **netsplice**
1746 Like **net**, but uses :manpage:`splice(2)` and
1747 :manpage:`vmsplice(2)` to map data and send/receive.
1748 This engine defines engine specific options.
1749
1750 **cpuio**
1751 Doesn't transfer any data, but burns CPU cycles according to the
1752 :option:`cpuload` and :option:`cpuchunks` options. Setting
9207a0cb 1753 :option:`cpuload`\=85 will cause that job to do nothing but burn 85%
71aa48eb 1754 of the CPU. In case of SMP machines, use :option:`numjobs`\=<nr_of_cpu>
f50fbdda 1755 to get desired CPU usage, as the cpuload only loads a
f80dba8d
MT
1756 single CPU at the desired rate. A job never finishes unless there is
1757 at least one non-cpuio job.
1758
1759 **guasi**
1760 The GUASI I/O engine is the Generic Userspace Asyncronous Syscall
1761 Interface approach to async I/O. See
1762
1763 http://www.xmailserver.org/guasi-lib.html
1764
1765 for more info on GUASI.
1766
1767 **rdma**
1768 The RDMA I/O engine supports both RDMA memory semantics
1769 (RDMA_WRITE/RDMA_READ) and channel semantics (Send/Recv) for the
609ac152
SB
1770 InfiniBand, RoCE and iWARP protocols. This engine defines engine
1771 specific options.
f80dba8d
MT
1772
1773 **falloc**
1774 I/O engine that does regular fallocate to simulate data transfer as
1775 fio ioengine.
1776
1777 DDIR_READ
1778 does fallocate(,mode = FALLOC_FL_KEEP_SIZE,).
1779
1780 DDIR_WRITE
1781 does fallocate(,mode = 0).
1782
1783 DDIR_TRIM
1784 does fallocate(,mode = FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE).
1785
761cd093
SW
1786 **ftruncate**
1787 I/O engine that sends :manpage:`ftruncate(2)` operations in response
1788 to write (DDIR_WRITE) events. Each ftruncate issued sets the file's
f50fbdda 1789 size to the current block offset. :option:`blocksize` is ignored.
761cd093 1790
f80dba8d
MT
1791 **e4defrag**
1792 I/O engine that does regular EXT4_IOC_MOVE_EXT ioctls to simulate
1793 defragment activity in request to DDIR_WRITE event.
1794
1795 **rbd**
1796 I/O engine supporting direct access to Ceph Rados Block Devices
1797 (RBD) via librbd without the need to use the kernel rbd driver. This
1798 ioengine defines engine specific options.
1799
1800 **gfapi**
ac8ca2af
SW
1801 Using GlusterFS libgfapi sync interface to direct access to
1802 GlusterFS volumes without having to go through FUSE. This ioengine
f80dba8d
MT
1803 defines engine specific options.
1804
1805 **gfapi_async**
ac8ca2af
SW
1806 Using GlusterFS libgfapi async interface to direct access to
1807 GlusterFS volumes without having to go through FUSE. This ioengine
f80dba8d
MT
1808 defines engine specific options.
1809
1810 **libhdfs**
f50fbdda 1811 Read and write through Hadoop (HDFS). The :option:`filename` option
f80dba8d
MT
1812 is used to specify host,port of the hdfs name-node to connect. This
1813 engine interprets offsets a little differently. In HDFS, files once
e25c0c91
SW
1814 created cannot be modified so random writes are not possible. To
1815 imitate this the libhdfs engine expects a bunch of small files to be
1816 created over HDFS and will randomly pick a file from them
1817 based on the offset generated by fio backend (see the example
f80dba8d 1818 job file to create such files, use ``rw=write`` option). Please
e25c0c91
SW
1819 note, it may be necessary to set environment variables to work
1820 with HDFS/libhdfs properly. Each job uses its own connection to
f80dba8d
MT
1821 HDFS.
1822
1823 **mtd**
1824 Read, write and erase an MTD character device (e.g.,
1825 :file:`/dev/mtd0`). Discards are treated as erases. Depending on the
1826 underlying device type, the I/O may have to go in a certain pattern,
1827 e.g., on NAND, writing sequentially to erase blocks and discarding
c298ee71 1828 before overwriting. The `trimwrite` mode works well for this
f80dba8d
MT
1829 constraint.
1830
1831 **pmemblk**
1832 Read and write using filesystem DAX to a file on a filesystem
1833 mounted with DAX on a persistent memory device through the NVML
1834 libpmemblk library.
1835
1836 **dev-dax**
1837 Read and write using device DAX to a persistent memory device (e.g.,
1838 /dev/dax0.0) through the NVML libpmem library.
1839
1840 **external**
1841 Prefix to specify loading an external I/O engine object file. Append
c60ebc45 1842 the engine filename, e.g. ``ioengine=external:/tmp/foo.o`` to load
d243fd6d
TK
1843 ioengine :file:`foo.o` in :file:`/tmp`. The path can be either
1844 absolute or relative. See :file:`engines/skeleton_external.c` for
1845 details of writing an external I/O engine.
f80dba8d 1846
1216cc5a 1847 **filecreate**
b71968b1 1848 Simply create the files and do no I/O to them. You still need to
1216cc5a 1849 set `filesize` so that all the accounting still occurs, but no
b71968b1 1850 actual I/O will be done other than creating the file.
f80dba8d 1851
ae0db592
TI
1852 **libpmem**
1853 Read and write using mmap I/O to a file on a filesystem
1854 mounted with DAX on a persistent memory device through the NVML
1855 libpmem library.
1856
f80dba8d
MT
1857I/O engine specific parameters
1858~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1859
1860In addition, there are some parameters which are only valid when a specific
f50fbdda
TK
1861:option:`ioengine` is in use. These are used identically to normal parameters,
1862with the caveat that when used on the command line, they must come after the
f80dba8d
MT
1863:option:`ioengine` that defines them is selected.
1864
1865.. option:: userspace_reap : [libaio]
1866
1867 Normally, with the libaio engine in use, fio will use the
1868 :manpage:`io_getevents(2)` system call to reap newly returned events. With
1869 this flag turned on, the AIO ring will be read directly from user-space to
1870 reap events. The reaping mode is only enabled when polling for a minimum of
c60ebc45 1871 0 events (e.g. when :option:`iodepth_batch_complete` `=0`).
f80dba8d 1872
9d25d068 1873.. option:: hipri : [pvsync2]
f80dba8d
MT
1874
1875 Set RWF_HIPRI on I/O, indicating to the kernel that it's of higher priority
1876 than normal.
1877
a0679ce5
SB
1878.. option:: hipri_percentage : [pvsync2]
1879
f50fbdda 1880 When hipri is set this determines the probability of a pvsync2 I/O being high
a0679ce5
SB
1881 priority. The default is 100%.
1882
f80dba8d
MT
1883.. option:: cpuload=int : [cpuio]
1884
da19cdb4
TK
1885 Attempt to use the specified percentage of CPU cycles. This is a mandatory
1886 option when using cpuio I/O engine.
f80dba8d
MT
1887
1888.. option:: cpuchunks=int : [cpuio]
1889
1890 Split the load into cycles of the given time. In microseconds.
1891
1892.. option:: exit_on_io_done=bool : [cpuio]
1893
1894 Detect when I/O threads are done, then exit.
1895
f80dba8d
MT
1896.. option:: namenode=str : [libhdfs]
1897
22413915 1898 The hostname or IP address of a HDFS cluster namenode to contact.
f80dba8d
MT
1899
1900.. option:: port=int
1901
f50fbdda
TK
1902 [libhdfs]
1903
1904 The listening port of the HFDS cluster namenode.
1905
f80dba8d
MT
1906 [netsplice], [net]
1907
1908 The TCP or UDP port to bind to or connect to. If this is used with
1909 :option:`numjobs` to spawn multiple instances of the same job type, then
1910 this will be the starting port number since fio will use a range of
1911 ports.
1912
609ac152
SB
1913 [rdma]
1914
1915 The port to use for RDMA-CM communication. This should be the same value
1916 on the client and the server side.
1917
1918.. option:: hostname=str : [netsplice] [net] [rdma]
f80dba8d 1919
609ac152
SB
1920 The hostname or IP address to use for TCP, UDP or RDMA-CM based I/O. If the job
1921 is a TCP listener or UDP reader, the hostname is not used and must be omitted
f50fbdda 1922 unless it is a valid UDP multicast address.
f80dba8d
MT
1923
1924.. option:: interface=str : [netsplice] [net]
1925
1926 The IP address of the network interface used to send or receive UDP
1927 multicast.
1928
1929.. option:: ttl=int : [netsplice] [net]
1930
1931 Time-to-live value for outgoing UDP multicast packets. Default: 1.
1932
1933.. option:: nodelay=bool : [netsplice] [net]
1934
1935 Set TCP_NODELAY on TCP connections.
1936
f50fbdda 1937.. option:: protocol=str, proto=str : [netsplice] [net]
f80dba8d
MT
1938
1939 The network protocol to use. Accepted values are:
1940
1941 **tcp**
1942 Transmission control protocol.
1943 **tcpv6**
1944 Transmission control protocol V6.
1945 **udp**
1946 User datagram protocol.
1947 **udpv6**
1948 User datagram protocol V6.
1949 **unix**
1950 UNIX domain socket.
1951
1952 When the protocol is TCP or UDP, the port must also be given, as well as the
1953 hostname if the job is a TCP listener or UDP reader. For unix sockets, the
f50fbdda 1954 normal :option:`filename` option should be used and the port is invalid.
f80dba8d 1955
e9184ec1 1956.. option:: listen : [netsplice] [net]
f80dba8d
MT
1957
1958 For TCP network connections, tell fio to listen for incoming connections
1959 rather than initiating an outgoing connection. The :option:`hostname` must
1960 be omitted if this option is used.
1961
e9184ec1 1962.. option:: pingpong : [netsplice] [net]
f80dba8d
MT
1963
1964 Normally a network writer will just continue writing data, and a network
1965 reader will just consume packages. If ``pingpong=1`` is set, a writer will
1966 send its normal payload to the reader, then wait for the reader to send the
1967 same payload back. This allows fio to measure network latencies. The
1968 submission and completion latencies then measure local time spent sending or
1969 receiving, and the completion latency measures how long it took for the
1970 other end to receive and send back. For UDP multicast traffic
1971 ``pingpong=1`` should only be set for a single reader when multiple readers
1972 are listening to the same address.
1973
e9184ec1 1974.. option:: window_size : [netsplice] [net]
f80dba8d
MT
1975
1976 Set the desired socket buffer size for the connection.
1977
e9184ec1 1978.. option:: mss : [netsplice] [net]
f80dba8d
MT
1979
1980 Set the TCP maximum segment size (TCP_MAXSEG).
1981
1982.. option:: donorname=str : [e4defrag]
1983
730bd7d9 1984 File will be used as a block donor (swap extents between files).
f80dba8d
MT
1985
1986.. option:: inplace=int : [e4defrag]
1987
1988 Configure donor file blocks allocation strategy:
1989
1990 **0**
1991 Default. Preallocate donor's file on init.
1992 **1**
2b455dbf 1993 Allocate space immediately inside defragment event, and free right
f80dba8d
MT
1994 after event.
1995
1996.. option:: clustername=str : [rbd]
1997
1998 Specifies the name of the Ceph cluster.
1999
2000.. option:: rbdname=str : [rbd]
2001
2002 Specifies the name of the RBD.
2003
2004.. option:: pool=str : [rbd]
2005
2006 Specifies the name of the Ceph pool containing RBD.
2007
2008.. option:: clientname=str : [rbd]
2009
2010 Specifies the username (without the 'client.' prefix) used to access the
2011 Ceph cluster. If the *clustername* is specified, the *clientname* shall be
2012 the full *type.id* string. If no type. prefix is given, fio will add
2013 'client.' by default.
2014
2015.. option:: skip_bad=bool : [mtd]
2016
2017 Skip operations against known bad blocks.
2018
2019.. option:: hdfsdirectory : [libhdfs]
2020
2021 libhdfs will create chunk in this HDFS directory.
2022
2023.. option:: chunk_size : [libhdfs]
2024
2b455dbf 2025 The size of the chunk to use for each file.
f80dba8d 2026
609ac152
SB
2027.. option:: verb=str : [rdma]
2028
2029 The RDMA verb to use on this side of the RDMA ioengine connection. Valid
2030 values are write, read, send and recv. These correspond to the equivalent
2031 RDMA verbs (e.g. write = rdma_write etc.). Note that this only needs to be
2032 specified on the client side of the connection. See the examples folder.
2033
2034.. option:: bindname=str : [rdma]
2035
2036 The name to use to bind the local RDMA-CM connection to a local RDMA device.
2037 This could be a hostname or an IPv4 or IPv6 address. On the server side this
2038 will be passed into the rdma_bind_addr() function and on the client site it
2039 will be used in the rdma_resolve_add() function. This can be useful when
2040 multiple paths exist between the client and the server or in certain loopback
2041 configurations.
f80dba8d
MT
2042
2043I/O depth
2044~~~~~~~~~
2045
2046.. option:: iodepth=int
2047
2048 Number of I/O units to keep in flight against the file. Note that
2049 increasing *iodepth* beyond 1 will not affect synchronous ioengines (except
c60ebc45 2050 for small degrees when :option:`verify_async` is in use). Even async
f80dba8d
MT
2051 engines may impose OS restrictions causing the desired depth not to be
2052 achieved. This may happen on Linux when using libaio and not setting
9207a0cb 2053 :option:`direct`\=1, since buffered I/O is not async on that OS. Keep an
f80dba8d
MT
2054 eye on the I/O depth distribution in the fio output to verify that the
2055 achieved depth is as expected. Default: 1.
2056
2057.. option:: iodepth_batch_submit=int, iodepth_batch=int
2058
2059 This defines how many pieces of I/O to submit at once. It defaults to 1
2060 which means that we submit each I/O as soon as it is available, but can be
2061 raised to submit bigger batches of I/O at the time. If it is set to 0 the
2062 :option:`iodepth` value will be used.
2063
2064.. option:: iodepth_batch_complete_min=int, iodepth_batch_complete=int
2065
2066 This defines how many pieces of I/O to retrieve at once. It defaults to 1
2067 which means that we'll ask for a minimum of 1 I/O in the retrieval process
2068 from the kernel. The I/O retrieval will go on until we hit the limit set by
2069 :option:`iodepth_low`. If this variable is set to 0, then fio will always
2070 check for completed events before queuing more I/O. This helps reduce I/O
2071 latency, at the cost of more retrieval system calls.
2072
2073.. option:: iodepth_batch_complete_max=int
2074
2075 This defines maximum pieces of I/O to retrieve at once. This variable should
9207a0cb 2076 be used along with :option:`iodepth_batch_complete_min`\=int variable,
f80dba8d 2077 specifying the range of min and max amount of I/O which should be
730bd7d9 2078 retrieved. By default it is equal to the :option:`iodepth_batch_complete_min`
f80dba8d
MT
2079 value.
2080
2081 Example #1::
2082
2083 iodepth_batch_complete_min=1
2084 iodepth_batch_complete_max=<iodepth>
2085
2086 which means that we will retrieve at least 1 I/O and up to the whole
2087 submitted queue depth. If none of I/O has been completed yet, we will wait.
2088
2089 Example #2::
2090
2091 iodepth_batch_complete_min=0
2092 iodepth_batch_complete_max=<iodepth>
2093
2094 which means that we can retrieve up to the whole submitted queue depth, but
2095 if none of I/O has been completed yet, we will NOT wait and immediately exit
2096 the system call. In this example we simply do polling.
2097
2098.. option:: iodepth_low=int
2099
2100 The low water mark indicating when to start filling the queue
2101 again. Defaults to the same as :option:`iodepth`, meaning that fio will
2102 attempt to keep the queue full at all times. If :option:`iodepth` is set to
c60ebc45 2103 e.g. 16 and *iodepth_low* is set to 4, then after fio has filled the queue of
f80dba8d
MT
2104 16 requests, it will let the depth drain down to 4 before starting to fill
2105 it again.
2106
997b5680
SW
2107.. option:: serialize_overlap=bool
2108
2109 Serialize in-flight I/Os that might otherwise cause or suffer from data races.
2110 When two or more I/Os are submitted simultaneously, there is no guarantee that
2111 the I/Os will be processed or completed in the submitted order. Further, if
2112 two or more of those I/Os are writes, any overlapping region between them can
2113 become indeterminate/undefined on certain storage. These issues can cause
2114 verification to fail erratically when at least one of the racing I/Os is
2115 changing data and the overlapping region has a non-zero size. Setting
2116 ``serialize_overlap`` tells fio to avoid provoking this behavior by explicitly
2117 serializing in-flight I/Os that have a non-zero overlap. Note that setting
ee21ebee 2118 this option can reduce both performance and the :option:`iodepth` achieved.
997b5680
SW
2119 Additionally this option does not work when :option:`io_submit_mode` is set to
2120 offload. Default: false.
2121
f80dba8d
MT
2122.. option:: io_submit_mode=str
2123
2124 This option controls how fio submits the I/O to the I/O engine. The default
2125 is `inline`, which means that the fio job threads submit and reap I/O
2126 directly. If set to `offload`, the job threads will offload I/O submission
2127 to a dedicated pool of I/O threads. This requires some coordination and thus
2128 has a bit of extra overhead, especially for lower queue depth I/O where it
2129 can increase latencies. The benefit is that fio can manage submission rates
2130 independently of the device completion rates. This avoids skewed latency
730bd7d9 2131 reporting if I/O gets backed up on the device side (the coordinated omission
f80dba8d
MT
2132 problem).
2133
2134
2135I/O rate
2136~~~~~~~~
2137
a881438b 2138.. option:: thinktime=time
f80dba8d 2139
f75ede1d
SW
2140 Stall the job for the specified period of time after an I/O has completed before issuing the
2141 next. May be used to simulate processing being done by an application.
947e0fe0 2142 When the unit is omitted, the value is interpreted in microseconds. See
f80dba8d
MT
2143 :option:`thinktime_blocks` and :option:`thinktime_spin`.
2144
a881438b 2145.. option:: thinktime_spin=time
f80dba8d
MT
2146
2147 Only valid if :option:`thinktime` is set - pretend to spend CPU time doing
2148 something with the data received, before falling back to sleeping for the
f75ede1d 2149 rest of the period specified by :option:`thinktime`. When the unit is
947e0fe0 2150 omitted, the value is interpreted in microseconds.
f80dba8d
MT
2151
2152.. option:: thinktime_blocks=int
2153
2154 Only valid if :option:`thinktime` is set - control how many blocks to issue,
f50fbdda
TK
2155 before waiting :option:`thinktime` usecs. If not set, defaults to 1 which will make
2156 fio wait :option:`thinktime` usecs after every block. This effectively makes any
f80dba8d 2157 queue depth setting redundant, since no more than 1 I/O will be queued
f50fbdda 2158 before we have to complete it and do our :option:`thinktime`. In other words, this
f80dba8d 2159 setting effectively caps the queue depth if the latter is larger.
71bfa161 2160
f80dba8d 2161.. option:: rate=int[,int][,int]
71bfa161 2162
f80dba8d
MT
2163 Cap the bandwidth used by this job. The number is in bytes/sec, the normal
2164 suffix rules apply. Comma-separated values may be specified for reads,
2165 writes, and trims as described in :option:`blocksize`.
71bfa161 2166
b25b3464
SW
2167 For example, using `rate=1m,500k` would limit reads to 1MiB/sec and writes to
2168 500KiB/sec. Capping only reads or writes can be done with `rate=,500k` or
2169 `rate=500k,` where the former will only limit writes (to 500KiB/sec) and the
2170 latter will only limit reads.
2171
f80dba8d 2172.. option:: rate_min=int[,int][,int]
71bfa161 2173
f80dba8d
MT
2174 Tell fio to do whatever it can to maintain at least this bandwidth. Failing
2175 to meet this requirement will cause the job to exit. Comma-separated values
2176 may be specified for reads, writes, and trims as described in
2177 :option:`blocksize`.
71bfa161 2178
f80dba8d 2179.. option:: rate_iops=int[,int][,int]
71bfa161 2180
f80dba8d
MT
2181 Cap the bandwidth to this number of IOPS. Basically the same as
2182 :option:`rate`, just specified independently of bandwidth. If the job is
2183 given a block size range instead of a fixed value, the smallest block size
2184 is used as the metric. Comma-separated values may be specified for reads,
2185 writes, and trims as described in :option:`blocksize`.
71bfa161 2186
f80dba8d 2187.. option:: rate_iops_min=int[,int][,int]
71bfa161 2188
f80dba8d
MT
2189 If fio doesn't meet this rate of I/O, it will cause the job to exit.
2190 Comma-separated values may be specified for reads, writes, and trims as
2191 described in :option:`blocksize`.
71bfa161 2192
f80dba8d 2193.. option:: rate_process=str
66c098b8 2194
f80dba8d
MT
2195 This option controls how fio manages rated I/O submissions. The default is
2196 `linear`, which submits I/O in a linear fashion with fixed delays between
c60ebc45 2197 I/Os that gets adjusted based on I/O completion rates. If this is set to
f80dba8d
MT
2198 `poisson`, fio will submit I/O based on a more real world random request
2199 flow, known as the Poisson process
2200 (https://en.wikipedia.org/wiki/Poisson_point_process). The lambda will be
2201 10^6 / IOPS for the given workload.
71bfa161
JA
2202
2203
f80dba8d
MT
2204I/O latency
2205~~~~~~~~~~~
71bfa161 2206
a881438b 2207.. option:: latency_target=time
71bfa161 2208
f80dba8d 2209 If set, fio will attempt to find the max performance point that the given
f75ede1d 2210 workload will run at while maintaining a latency below this target. When
947e0fe0 2211 the unit is omitted, the value is interpreted in microseconds. See
f75ede1d 2212 :option:`latency_window` and :option:`latency_percentile`.
71bfa161 2213
a881438b 2214.. option:: latency_window=time
71bfa161 2215
f80dba8d 2216 Used with :option:`latency_target` to specify the sample window that the job
f75ede1d 2217 is run at varying queue depths to test the performance. When the unit is
947e0fe0 2218 omitted, the value is interpreted in microseconds.
b4692828 2219
f80dba8d 2220.. option:: latency_percentile=float
71bfa161 2221
c60ebc45 2222 The percentage of I/Os that must fall within the criteria specified by
f80dba8d 2223 :option:`latency_target` and :option:`latency_window`. If not set, this
c60ebc45 2224 defaults to 100.0, meaning that all I/Os must be equal or below to the value
f80dba8d 2225 set by :option:`latency_target`.
71bfa161 2226
a881438b 2227.. option:: max_latency=time
71bfa161 2228
f75ede1d 2229 If set, fio will exit the job with an ETIMEDOUT error if it exceeds this
947e0fe0 2230 maximum latency. When the unit is omitted, the value is interpreted in
f75ede1d 2231 microseconds.
71bfa161 2232
f80dba8d 2233.. option:: rate_cycle=int
71bfa161 2234
f80dba8d 2235 Average bandwidth for :option:`rate` and :option:`rate_min` over this number
a47b697c 2236 of milliseconds. Defaults to 1000.
71bfa161 2237
71bfa161 2238
f80dba8d
MT
2239I/O replay
2240~~~~~~~~~~
71bfa161 2241
f80dba8d 2242.. option:: write_iolog=str
c2b1e753 2243
f80dba8d
MT
2244 Write the issued I/O patterns to the specified file. See
2245 :option:`read_iolog`. Specify a separate file for each job, otherwise the
2246 iologs will be interspersed and the file may be corrupt.
c2b1e753 2247
f80dba8d 2248.. option:: read_iolog=str
71bfa161 2249
22413915 2250 Open an iolog with the specified filename and replay the I/O patterns it
f80dba8d
MT
2251 contains. This can be used to store a workload and replay it sometime
2252 later. The iolog given may also be a blktrace binary file, which allows fio
2253 to replay a workload captured by :command:`blktrace`. See
2254 :manpage:`blktrace(8)` for how to capture such logging data. For blktrace
2255 replay, the file needs to be turned into a blkparse binary data file first
2256 (``blkparse <device> -o /dev/null -d file_for_fio.bin``).
71bfa161 2257
589e88b7 2258.. option:: replay_no_stall=bool
71bfa161 2259
f80dba8d 2260 When replaying I/O with :option:`read_iolog` the default behavior is to
22413915 2261 attempt to respect the timestamps within the log and replay them with the
f80dba8d
MT
2262 appropriate delay between IOPS. By setting this variable fio will not
2263 respect the timestamps and attempt to replay them as fast as possible while
2264 still respecting ordering. The result is the same I/O pattern to a given
2265 device, but different timings.
71bfa161 2266
f80dba8d 2267.. option:: replay_redirect=str
b4692828 2268
f80dba8d
MT
2269 While replaying I/O patterns using :option:`read_iolog` the default behavior
2270 is to replay the IOPS onto the major/minor device that each IOP was recorded
2271 from. This is sometimes undesirable because on a different machine those
2272 major/minor numbers can map to a different device. Changing hardware on the
2273 same system can also result in a different major/minor mapping.
730bd7d9 2274 ``replay_redirect`` causes all I/Os to be replayed onto the single specified
f80dba8d 2275 device regardless of the device it was recorded
9207a0cb 2276 from. i.e. :option:`replay_redirect`\= :file:`/dev/sdc` would cause all I/O
f80dba8d
MT
2277 in the blktrace or iolog to be replayed onto :file:`/dev/sdc`. This means
2278 multiple devices will be replayed onto a single device, if the trace
2279 contains multiple devices. If you want multiple devices to be replayed
2280 concurrently to multiple redirected devices you must blkparse your trace
2281 into separate traces and replay them with independent fio invocations.
2282 Unfortunately this also breaks the strict time ordering between multiple
2283 device accesses.
71bfa161 2284
f80dba8d 2285.. option:: replay_align=int
74929ac2 2286
f80dba8d
MT
2287 Force alignment of I/O offsets and lengths in a trace to this power of 2
2288 value.
3c54bc46 2289
f80dba8d 2290.. option:: replay_scale=int
3c54bc46 2291
f80dba8d 2292 Scale sector offsets down by this factor when replaying traces.
3c54bc46 2293
3c54bc46 2294
f80dba8d
MT
2295Threads, processes and job synchronization
2296~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3c54bc46 2297
f80dba8d 2298.. option:: thread
3c54bc46 2299
730bd7d9
SW
2300 Fio defaults to creating jobs by using fork, however if this option is
2301 given, fio will create jobs by using POSIX Threads' function
2302 :manpage:`pthread_create(3)` to create threads instead.
71bfa161 2303
f80dba8d 2304.. option:: wait_for=str
74929ac2 2305
730bd7d9
SW
2306 If set, the current job won't be started until all workers of the specified
2307 waitee job are done.
74929ac2 2308
f80dba8d
MT
2309 ``wait_for`` operates on the job name basis, so there are a few
2310 limitations. First, the waitee must be defined prior to the waiter job
2311 (meaning no forward references). Second, if a job is being referenced as a
2312 waitee, it must have a unique name (no duplicate waitees).
74929ac2 2313
f80dba8d 2314.. option:: nice=int
892a6ffc 2315
f80dba8d 2316 Run the job with the given nice value. See man :manpage:`nice(2)`.
892a6ffc 2317
f80dba8d
MT
2318 On Windows, values less than -15 set the process class to "High"; -1 through
2319 -15 set "Above Normal"; 1 through 15 "Below Normal"; and above 15 "Idle"
2320 priority class.
74929ac2 2321
f80dba8d 2322.. option:: prio=int
71bfa161 2323
f80dba8d
MT
2324 Set the I/O priority value of this job. Linux limits us to a positive value
2325 between 0 and 7, with 0 being the highest. See man
2326 :manpage:`ionice(1)`. Refer to an appropriate manpage for other operating
2327 systems since meaning of priority may differ.
71bfa161 2328
f80dba8d 2329.. option:: prioclass=int
d59aa780 2330
f80dba8d 2331 Set the I/O priority class. See man :manpage:`ionice(1)`.
d59aa780 2332
f80dba8d 2333.. option:: cpumask=int
71bfa161 2334
22413915
SW
2335 Set the CPU affinity of this job. The parameter given is a bit mask of
2336 allowed CPUs the job may run on. So if you want the allowed CPUs to be 1
f80dba8d
MT
2337 and 5, you would pass the decimal value of (1 << 1 | 1 << 5), or 34. See man
2338 :manpage:`sched_setaffinity(2)`. This may not work on all supported
2339 operating systems or kernel versions. This option doesn't work well for a
2340 higher CPU count than what you can store in an integer mask, so it can only
2341 control cpus 1-32. For boxes with larger CPU counts, use
2342 :option:`cpus_allowed`.
6d500c2e 2343
f80dba8d 2344.. option:: cpus_allowed=str
6d500c2e 2345
730bd7d9
SW
2346 Controls the same options as :option:`cpumask`, but accepts a textual
2347 specification of the permitted CPUs instead. So to use CPUs 1 and 5 you
2348 would specify ``cpus_allowed=1,5``. This option also allows a range of CPUs
2349 to be specified -- say you wanted a binding to CPUs 1, 5, and 8 to 15, you
2350 would set ``cpus_allowed=1,5,8-15``.
6d500c2e 2351
f80dba8d 2352.. option:: cpus_allowed_policy=str
6d500c2e 2353
f80dba8d 2354 Set the policy of how fio distributes the CPUs specified by
730bd7d9 2355 :option:`cpus_allowed` or :option:`cpumask`. Two policies are supported:
6d500c2e 2356
f80dba8d
MT
2357 **shared**
2358 All jobs will share the CPU set specified.
2359 **split**
2360 Each job will get a unique CPU from the CPU set.
6d500c2e 2361
22413915 2362 **shared** is the default behavior, if the option isn't specified. If
f80dba8d
MT
2363 **split** is specified, then fio will will assign one cpu per job. If not
2364 enough CPUs are given for the jobs listed, then fio will roundrobin the CPUs
2365 in the set.
6d500c2e 2366
f80dba8d 2367.. option:: numa_cpu_nodes=str
6d500c2e 2368
f80dba8d
MT
2369 Set this job running on specified NUMA nodes' CPUs. The arguments allow
2370 comma delimited list of cpu numbers, A-B ranges, or `all`. Note, to enable
ac8ca2af 2371 NUMA options support, fio must be built on a system with libnuma-dev(el)
f80dba8d 2372 installed.
61b9861d 2373
f80dba8d 2374.. option:: numa_mem_policy=str
61b9861d 2375
f80dba8d
MT
2376 Set this job's memory policy and corresponding NUMA nodes. Format of the
2377 arguments::
5c94b008 2378
f80dba8d 2379 <mode>[:<nodelist>]
ce35b1ec 2380
730bd7d9
SW
2381 ``mode`` is one of the following memory poicies: ``default``, ``prefer``,
2382 ``bind``, ``interleave`` or ``local``. For ``default`` and ``local`` memory
2383 policies, no node needs to be specified. For ``prefer``, only one node is
2384 allowed. For ``bind`` and ``interleave`` the ``nodelist`` may be as
2385 follows: a comma delimited list of numbers, A-B ranges, or `all`.
71bfa161 2386
f80dba8d 2387.. option:: cgroup=str
390b1537 2388
f80dba8d
MT
2389 Add job to this control group. If it doesn't exist, it will be created. The
2390 system must have a mounted cgroup blkio mount point for this to work. If
2391 your system doesn't have it mounted, you can do so with::
5af1c6f3 2392
f80dba8d 2393 # mount -t cgroup -o blkio none /cgroup
5af1c6f3 2394
f80dba8d 2395.. option:: cgroup_weight=int
5af1c6f3 2396
f80dba8d
MT
2397 Set the weight of the cgroup to this value. See the documentation that comes
2398 with the kernel, allowed values are in the range of 100..1000.
a086c257 2399
f80dba8d 2400.. option:: cgroup_nodelete=bool
8c07860d 2401
f80dba8d
MT
2402 Normally fio will delete the cgroups it has created after the job
2403 completion. To override this behavior and to leave cgroups around after the
2404 job completion, set ``cgroup_nodelete=1``. This can be useful if one wants
2405 to inspect various cgroup files after job completion. Default: false.
8c07860d 2406
f80dba8d 2407.. option:: flow_id=int
8c07860d 2408
f80dba8d
MT
2409 The ID of the flow. If not specified, it defaults to being a global
2410 flow. See :option:`flow`.
1907dbc6 2411
f80dba8d 2412.. option:: flow=int
71bfa161 2413
f80dba8d
MT
2414 Weight in token-based flow control. If this value is used, then there is a
2415 'flow counter' which is used to regulate the proportion of activity between
2416 two or more jobs. Fio attempts to keep this flow counter near zero. The
2417 ``flow`` parameter stands for how much should be added or subtracted to the
2418 flow counter on each iteration of the main I/O loop. That is, if one job has
2419 ``flow=8`` and another job has ``flow=-1``, then there will be a roughly 1:8
2420 ratio in how much one runs vs the other.
71bfa161 2421
f80dba8d 2422.. option:: flow_watermark=int
a31041ea 2423
f80dba8d
MT
2424 The maximum value that the absolute value of the flow counter is allowed to
2425 reach before the job must wait for a lower value of the counter.
82407585 2426
f80dba8d 2427.. option:: flow_sleep=int
82407585 2428
f80dba8d
MT
2429 The period of time, in microseconds, to wait after the flow watermark has
2430 been exceeded before retrying operations.
82407585 2431
f80dba8d 2432.. option:: stonewall, wait_for_previous
82407585 2433
f80dba8d
MT
2434 Wait for preceding jobs in the job file to exit, before starting this
2435 one. Can be used to insert serialization points in the job file. A stone
2436 wall also implies starting a new reporting group, see
2437 :option:`group_reporting`.
2438
2439.. option:: exitall
2440
730bd7d9
SW
2441 By default, fio will continue running all other jobs when one job finishes
2442 but sometimes this is not the desired action. Setting ``exitall`` will
2443 instead make fio terminate all other jobs when one job finishes.
f80dba8d
MT
2444
2445.. option:: exec_prerun=str
2446
2447 Before running this job, issue the command specified through
2448 :manpage:`system(3)`. Output is redirected in a file called
2449 :file:`jobname.prerun.txt`.
2450
2451.. option:: exec_postrun=str
2452
2453 After the job completes, issue the command specified though
2454 :manpage:`system(3)`. Output is redirected in a file called
2455 :file:`jobname.postrun.txt`.
2456
2457.. option:: uid=int
2458
2459 Instead of running as the invoking user, set the user ID to this value
2460 before the thread/process does any work.
2461
2462.. option:: gid=int
2463
2464 Set group ID, see :option:`uid`.
2465
2466
2467Verification
2468~~~~~~~~~~~~
2469
2470.. option:: verify_only
2471
2472 Do not perform specified workload, only verify data still matches previous
2473 invocation of this workload. This option allows one to check data multiple
2474 times at a later date without overwriting it. This option makes sense only
2475 for workloads that write data, and does not support workloads with the
2476 :option:`time_based` option set.
2477
2478.. option:: do_verify=bool
2479
2480 Run the verify phase after a write phase. Only valid if :option:`verify` is
2481 set. Default: true.
2482
2483.. option:: verify=str
2484
2485 If writing to a file, fio can verify the file contents after each iteration
2486 of the job. Each verification method also implies verification of special
2487 header, which is written to the beginning of each block. This header also
2488 includes meta information, like offset of the block, block number, timestamp
2489 when block was written, etc. :option:`verify` can be combined with
2490 :option:`verify_pattern` option. The allowed values are:
2491
2492 **md5**
2493 Use an md5 sum of the data area and store it in the header of
2494 each block.
2495
2496 **crc64**
2497 Use an experimental crc64 sum of the data area and store it in the
2498 header of each block.
2499
2500 **crc32c**
a5896300
SW
2501 Use a crc32c sum of the data area and store it in the header of
2502 each block. This will automatically use hardware acceleration
2503 (e.g. SSE4.2 on an x86 or CRC crypto extensions on ARM64) but will
2504 fall back to software crc32c if none is found. Generally the
2505 fatest checksum fio supports when hardware accelerated.
f80dba8d
MT
2506
2507 **crc32c-intel**
a5896300 2508 Synonym for crc32c.
f80dba8d
MT
2509
2510 **crc32**
2511 Use a crc32 sum of the data area and store it in the header of each
2512 block.
2513
2514 **crc16**
2515 Use a crc16 sum of the data area and store it in the header of each
2516 block.
2517
2518 **crc7**
2519 Use a crc7 sum of the data area and store it in the header of each
2520 block.
2521
2522 **xxhash**
2523 Use xxhash as the checksum function. Generally the fastest software
2524 checksum that fio supports.
2525
2526 **sha512**
2527 Use sha512 as the checksum function.
2528
2529 **sha256**
2530 Use sha256 as the checksum function.
2531
2532 **sha1**
2533 Use optimized sha1 as the checksum function.
82407585 2534
ae3a5acc
JA
2535 **sha3-224**
2536 Use optimized sha3-224 as the checksum function.
2537
2538 **sha3-256**
2539 Use optimized sha3-256 as the checksum function.
2540
2541 **sha3-384**
2542 Use optimized sha3-384 as the checksum function.
2543
2544 **sha3-512**
2545 Use optimized sha3-512 as the checksum function.
2546
f80dba8d
MT
2547 **meta**
2548 This option is deprecated, since now meta information is included in
2549 generic verification header and meta verification happens by
2550 default. For detailed information see the description of the
2551 :option:`verify` setting. This option is kept because of
2552 compatibility's sake with old configurations. Do not use it.
2553
2554 **pattern**
2555 Verify a strict pattern. Normally fio includes a header with some
2556 basic information and checksumming, but if this option is set, only
2557 the specific pattern set with :option:`verify_pattern` is verified.
2558
2559 **null**
2560 Only pretend to verify. Useful for testing internals with
9207a0cb 2561 :option:`ioengine`\=null, not for much else.
f80dba8d
MT
2562
2563 This option can be used for repeated burn-in tests of a system to make sure
2564 that the written data is also correctly read back. If the data direction
2565 given is a read or random read, fio will assume that it should verify a
2566 previously written file. If the data direction includes any form of write,
2567 the verify will be of the newly written data.
2568
2569.. option:: verifysort=bool
2570
2571 If true, fio will sort written verify blocks when it deems it faster to read
2572 them back in a sorted manner. This is often the case when overwriting an
2573 existing file, since the blocks are already laid out in the file system. You
2574 can ignore this option unless doing huge amounts of really fast I/O where
2575 the red-black tree sorting CPU time becomes significant. Default: true.
2576
2577.. option:: verifysort_nr=int
2578
2b455dbf 2579 Pre-load and sort verify blocks for a read workload.
f80dba8d
MT
2580
2581.. option:: verify_offset=int
2582
2583 Swap the verification header with data somewhere else in the block before
2584 writing. It is swapped back before verifying.
2585
2586.. option:: verify_interval=int
2587
2588 Write the verification header at a finer granularity than the
2589 :option:`blocksize`. It will be written for chunks the size of
2590 ``verify_interval``. :option:`blocksize` should divide this evenly.
2591
2592.. option:: verify_pattern=str
2593
2594 If set, fio will fill the I/O buffers with this pattern. Fio defaults to
2595 filling with totally random bytes, but sometimes it's interesting to fill
2596 with a known pattern for I/O verification purposes. Depending on the width
730bd7d9 2597 of the pattern, fio will fill 1/2/3/4 bytes of the buffer at the time (it can
f80dba8d
MT
2598 be either a decimal or a hex number). The ``verify_pattern`` if larger than
2599 a 32-bit quantity has to be a hex number that starts with either "0x" or
2600 "0X". Use with :option:`verify`. Also, ``verify_pattern`` supports %o
2601 format, which means that for each block offset will be written and then
2602 verified back, e.g.::
61b9861d
RP
2603
2604 verify_pattern=%o
2605
f80dba8d
MT
2606 Or use combination of everything::
2607
61b9861d 2608 verify_pattern=0xff%o"abcd"-12
e28218f3 2609
f80dba8d
MT
2610.. option:: verify_fatal=bool
2611
2612 Normally fio will keep checking the entire contents before quitting on a
2613 block verification failure. If this option is set, fio will exit the job on
2614 the first observed failure. Default: false.
2615
2616.. option:: verify_dump=bool
2617
2618 If set, dump the contents of both the original data block and the data block
2619 we read off disk to files. This allows later analysis to inspect just what
2620 kind of data corruption occurred. Off by default.
2621
2622.. option:: verify_async=int
2623
2624 Fio will normally verify I/O inline from the submitting thread. This option
2625 takes an integer describing how many async offload threads to create for I/O
2626 verification instead, causing fio to offload the duty of verifying I/O
2627 contents to one or more separate threads. If using this offload option, even
2628 sync I/O engines can benefit from using an :option:`iodepth` setting higher
2629 than 1, as it allows them to have I/O in flight while verifies are running.
d7e6ea1c 2630 Defaults to 0 async threads, i.e. verification is not asynchronous.
f80dba8d
MT
2631
2632.. option:: verify_async_cpus=str
2633
2634 Tell fio to set the given CPU affinity on the async I/O verification
2635 threads. See :option:`cpus_allowed` for the format used.
2636
2637.. option:: verify_backlog=int
2638
2639 Fio will normally verify the written contents of a job that utilizes verify
2640 once that job has completed. In other words, everything is written then
2641 everything is read back and verified. You may want to verify continually
2642 instead for a variety of reasons. Fio stores the meta data associated with
2643 an I/O block in memory, so for large verify workloads, quite a bit of memory
2644 would be used up holding this meta data. If this option is enabled, fio will
2645 write only N blocks before verifying these blocks.
2646
2647.. option:: verify_backlog_batch=int
2648
2649 Control how many blocks fio will verify if :option:`verify_backlog` is
2650 set. If not set, will default to the value of :option:`verify_backlog`
2651 (meaning the entire queue is read back and verified). If
2652 ``verify_backlog_batch`` is less than :option:`verify_backlog` then not all
2653 blocks will be verified, if ``verify_backlog_batch`` is larger than
2654 :option:`verify_backlog`, some blocks will be verified more than once.
2655
2656.. option:: verify_state_save=bool
2657
2658 When a job exits during the write phase of a verify workload, save its
2659 current state. This allows fio to replay up until that point, if the verify
2660 state is loaded for the verify read phase. The format of the filename is,
2661 roughly::
2662
f50fbdda 2663 <type>-<jobname>-<jobindex>-verify.state.
f80dba8d
MT
2664
2665 <type> is "local" for a local run, "sock" for a client/server socket
2666 connection, and "ip" (192.168.0.1, for instance) for a networked
d7e6ea1c 2667 client/server connection. Defaults to true.
f80dba8d
MT
2668
2669.. option:: verify_state_load=bool
2670
2671 If a verify termination trigger was used, fio stores the current write state
2672 of each thread. This can be used at verification time so that fio knows how
2673 far it should verify. Without this information, fio will run a full
a47b697c
SW
2674 verification pass, according to the settings in the job file used. Default
2675 false.
f80dba8d
MT
2676
2677.. option:: trim_percentage=int
2678
2679 Number of verify blocks to discard/trim.
2680
2681.. option:: trim_verify_zero=bool
2682
22413915 2683 Verify that trim/discarded blocks are returned as zeros.
f80dba8d
MT
2684
2685.. option:: trim_backlog=int
2686
5cfd1e9a 2687 Trim after this number of blocks are written.
f80dba8d
MT
2688
2689.. option:: trim_backlog_batch=int
2690
2691 Trim this number of I/O blocks.
2692
2693.. option:: experimental_verify=bool
2694
2695 Enable experimental verification.
2696
f80dba8d
MT
2697Steady state
2698~~~~~~~~~~~~
2699
2700.. option:: steadystate=str:float, ss=str:float
2701
2702 Define the criterion and limit for assessing steady state performance. The
2703 first parameter designates the criterion whereas the second parameter sets
2704 the threshold. When the criterion falls below the threshold for the
2705 specified duration, the job will stop. For example, `iops_slope:0.1%` will
2706 direct fio to terminate the job when the least squares regression slope
2707 falls below 0.1% of the mean IOPS. If :option:`group_reporting` is enabled
2708 this will apply to all jobs in the group. Below is the list of available
2709 steady state assessment criteria. All assessments are carried out using only
2710 data from the rolling collection window. Threshold limits can be expressed
2711 as a fixed value or as a percentage of the mean in the collection window.
2712
2713 **iops**
2714 Collect IOPS data. Stop the job if all individual IOPS measurements
2715 are within the specified limit of the mean IOPS (e.g., ``iops:2``
2716 means that all individual IOPS values must be within 2 of the mean,
2717 whereas ``iops:0.2%`` means that all individual IOPS values must be
2718 within 0.2% of the mean IOPS to terminate the job).
2719
2720 **iops_slope**
2721 Collect IOPS data and calculate the least squares regression
2722 slope. Stop the job if the slope falls below the specified limit.
2723
2724 **bw**
2725 Collect bandwidth data. Stop the job if all individual bandwidth
2726 measurements are within the specified limit of the mean bandwidth.
2727
2728 **bw_slope**
2729 Collect bandwidth data and calculate the least squares regression
2730 slope. Stop the job if the slope falls below the specified limit.
2731
2732.. option:: steadystate_duration=time, ss_dur=time
2733
2734 A rolling window of this duration will be used to judge whether steady state
2735 has been reached. Data will be collected once per second. The default is 0
f75ede1d 2736 which disables steady state detection. When the unit is omitted, the
947e0fe0 2737 value is interpreted in seconds.
f80dba8d
MT
2738
2739.. option:: steadystate_ramp_time=time, ss_ramp=time
2740
2741 Allow the job to run for the specified duration before beginning data
2742 collection for checking the steady state job termination criterion. The
947e0fe0 2743 default is 0. When the unit is omitted, the value is interpreted in seconds.
f80dba8d
MT
2744
2745
2746Measurements and reporting
2747~~~~~~~~~~~~~~~~~~~~~~~~~~
2748
2749.. option:: per_job_logs=bool
2750
2751 If set, this generates bw/clat/iops log with per file private filenames. If
2752 not set, jobs with identical names will share the log filename. Default:
2753 true.
2754
2755.. option:: group_reporting
2756
2757 It may sometimes be interesting to display statistics for groups of jobs as
2758 a whole instead of for each individual job. This is especially true if
2759 :option:`numjobs` is used; looking at individual thread/process output
2760 quickly becomes unwieldy. To see the final report per-group instead of
2761 per-job, use :option:`group_reporting`. Jobs in a file will be part of the
2762 same reporting group, unless if separated by a :option:`stonewall`, or by
2763 using :option:`new_group`.
2764
2765.. option:: new_group
2766
2767 Start a new reporting group. See: :option:`group_reporting`. If not given,
2768 all jobs in a file will be part of the same reporting group, unless
2769 separated by a :option:`stonewall`.
2770
589e88b7 2771.. option:: stats=bool
8243be59
JA
2772
2773 By default, fio collects and shows final output results for all jobs
2774 that run. If this option is set to 0, then fio will ignore it in
2775 the final stat output.
2776
f80dba8d
MT
2777.. option:: write_bw_log=str
2778
2779 If given, write a bandwidth log for this job. Can be used to store data of
074f0817 2780 the bandwidth of the jobs in their lifetime.
f80dba8d 2781
074f0817
SW
2782 If no str argument is given, the default filename of
2783 :file:`jobname_type.x.log` is used. Even when the argument is given, fio
2784 will still append the type of log. So if one specifies::
2785
2786 write_bw_log=foo
f80dba8d 2787
074f0817
SW
2788 The actual log name will be :file:`foo_bw.x.log` where `x` is the index
2789 of the job (`1..N`, where `N` is the number of jobs). If
2790 :option:`per_job_logs` is false, then the filename will not include the
2791 `.x` job index.
e3cedca7 2792
074f0817
SW
2793 The included :command:`fio_generate_plots` script uses :command:`gnuplot` to turn these
2794 text files into nice graphs. See `Log File Formats`_ for how data is
2795 structured within the file.
2796
2797.. option:: write_lat_log=str
e3cedca7 2798
074f0817 2799 Same as :option:`write_bw_log`, except this option creates I/O
77b7e675
SW
2800 submission (e.g., :file:`name_slat.x.log`), completion (e.g.,
2801 :file:`name_clat.x.log`), and total (e.g., :file:`name_lat.x.log`)
074f0817
SW
2802 latency files instead. See :option:`write_bw_log` for details about
2803 the filename format and `Log File Formats`_ for how data is structured
2804 within the files.
be4ecfdf 2805
f80dba8d 2806.. option:: write_hist_log=str
06842027 2807
074f0817 2808 Same as :option:`write_bw_log` but writes an I/O completion latency
77b7e675 2809 histogram file (e.g., :file:`name_hist.x.log`) instead. Note that this
074f0817
SW
2810 file will be empty unless :option:`log_hist_msec` has also been set.
2811 See :option:`write_bw_log` for details about the filename format and
2812 `Log File Formats`_ for how data is structured within the file.
06842027 2813
f80dba8d 2814.. option:: write_iops_log=str
06842027 2815
074f0817 2816 Same as :option:`write_bw_log`, but writes an IOPS file (e.g.
77b7e675 2817 :file:`name_iops.x.log`) instead. See :option:`write_bw_log` for
074f0817
SW
2818 details about the filename format and `Log File Formats`_ for how data
2819 is structured within the file.
06842027 2820
f80dba8d 2821.. option:: log_avg_msec=int
06842027 2822
f80dba8d
MT
2823 By default, fio will log an entry in the iops, latency, or bw log for every
2824 I/O that completes. When writing to the disk log, that can quickly grow to a
2825 very large size. Setting this option makes fio average the each log entry
2826 over the specified period of time, reducing the resolution of the log. See
2827 :option:`log_max_value` as well. Defaults to 0, logging all entries.
6fc82095 2828 Also see `Log File Formats`_.
06842027 2829
f80dba8d 2830.. option:: log_hist_msec=int
06842027 2831
f80dba8d
MT
2832 Same as :option:`log_avg_msec`, but logs entries for completion latency
2833 histograms. Computing latency percentiles from averages of intervals using
c60ebc45 2834 :option:`log_avg_msec` is inaccurate. Setting this option makes fio log
f80dba8d
MT
2835 histogram entries over the specified period of time, reducing log sizes for
2836 high IOPS devices while retaining percentile accuracy. See
074f0817
SW
2837 :option:`log_hist_coarseness` and :option:`write_hist_log` as well.
2838 Defaults to 0, meaning histogram logging is disabled.
06842027 2839
f80dba8d 2840.. option:: log_hist_coarseness=int
06842027 2841
f80dba8d
MT
2842 Integer ranging from 0 to 6, defining the coarseness of the resolution of
2843 the histogram logs enabled with :option:`log_hist_msec`. For each increment
2844 in coarseness, fio outputs half as many bins. Defaults to 0, for which
074f0817
SW
2845 histogram logs contain 1216 latency bins. See :option:`write_hist_log`
2846 and `Log File Formats`_.
8b28bd41 2847
f80dba8d 2848.. option:: log_max_value=bool
66c098b8 2849
f80dba8d
MT
2850 If :option:`log_avg_msec` is set, fio logs the average over that window. If
2851 you instead want to log the maximum value, set this option to 1. Defaults to
2852 0, meaning that averaged values are logged.
a696fa2a 2853
589e88b7 2854.. option:: log_offset=bool
a696fa2a 2855
f80dba8d 2856 If this is set, the iolog options will include the byte offset for the I/O
5a83478f
SW
2857 entry as well as the other data values. Defaults to 0 meaning that
2858 offsets are not present in logs. Also see `Log File Formats`_.
71bfa161 2859
f80dba8d 2860.. option:: log_compression=int
7de87099 2861
f80dba8d
MT
2862 If this is set, fio will compress the I/O logs as it goes, to keep the
2863 memory footprint lower. When a log reaches the specified size, that chunk is
2864 removed and compressed in the background. Given that I/O logs are fairly
2865 highly compressible, this yields a nice memory savings for longer runs. The
2866 downside is that the compression will consume some background CPU cycles, so
2867 it may impact the run. This, however, is also true if the logging ends up
2868 consuming most of the system memory. So pick your poison. The I/O logs are
2869 saved normally at the end of a run, by decompressing the chunks and storing
2870 them in the specified log file. This feature depends on the availability of
2871 zlib.
e0b0d892 2872
f80dba8d 2873.. option:: log_compression_cpus=str
e0b0d892 2874
f80dba8d
MT
2875 Define the set of CPUs that are allowed to handle online log compression for
2876 the I/O jobs. This can provide better isolation between performance
2877 sensitive jobs, and background compression work.
9e684a49 2878
f80dba8d 2879.. option:: log_store_compressed=bool
9e684a49 2880
f80dba8d
MT
2881 If set, fio will store the log files in a compressed format. They can be
2882 decompressed with fio, using the :option:`--inflate-log` command line
2883 parameter. The files will be stored with a :file:`.fz` suffix.
9e684a49 2884
f80dba8d 2885.. option:: log_unix_epoch=bool
9e684a49 2886
f80dba8d
MT
2887 If set, fio will log Unix timestamps to the log files produced by enabling
2888 write_type_log for each log type, instead of the default zero-based
2889 timestamps.
2890
2891.. option:: block_error_percentiles=bool
2892
2893 If set, record errors in trim block-sized units from writes and trims and
2894 output a histogram of how many trims it took to get to errors, and what kind
2895 of error was encountered.
2896
2897.. option:: bwavgtime=int
2898
2899 Average the calculated bandwidth over the given time. Value is specified in
2900 milliseconds. If the job also does bandwidth logging through
2901 :option:`write_bw_log`, then the minimum of this option and
2902 :option:`log_avg_msec` will be used. Default: 500ms.
2903
2904.. option:: iopsavgtime=int
2905
2906 Average the calculated IOPS over the given time. Value is specified in
2907 milliseconds. If the job also does IOPS logging through
2908 :option:`write_iops_log`, then the minimum of this option and
2909 :option:`log_avg_msec` will be used. Default: 500ms.
2910
2911.. option:: disk_util=bool
2912
2913 Generate disk utilization statistics, if the platform supports it.
2914 Default: true.
2915
2916.. option:: disable_lat=bool
2917
2918 Disable measurements of total latency numbers. Useful only for cutting back
2919 the number of calls to :manpage:`gettimeofday(2)`, as that does impact
2920 performance at really high IOPS rates. Note that to really get rid of a
2921 large amount of these calls, this option must be used with
f75ede1d 2922 :option:`disable_slat` and :option:`disable_bw_measurement` as well.
f80dba8d
MT
2923
2924.. option:: disable_clat=bool
2925
2926 Disable measurements of completion latency numbers. See
2927 :option:`disable_lat`.
2928
2929.. option:: disable_slat=bool
2930
2931 Disable measurements of submission latency numbers. See
f50fbdda 2932 :option:`disable_lat`.
f80dba8d 2933
f75ede1d 2934.. option:: disable_bw_measurement=bool, disable_bw=bool
f80dba8d
MT
2935
2936 Disable measurements of throughput/bandwidth numbers. See
2937 :option:`disable_lat`.
2938
2939.. option:: clat_percentiles=bool
2940
b599759b
JA
2941 Enable the reporting of percentiles of completion latencies. This
2942 option is mutually exclusive with :option:`lat_percentiles`.
2943
2944.. option:: lat_percentiles=bool
2945
b71968b1 2946 Enable the reporting of percentiles of I/O latencies. This is similar
b599759b
JA
2947 to :option:`clat_percentiles`, except that this includes the
2948 submission latency. This option is mutually exclusive with
2949 :option:`clat_percentiles`.
f80dba8d
MT
2950
2951.. option:: percentile_list=float_list
2952
c32ba107
JA
2953 Overwrite the default list of percentiles for completion latencies and
2954 the block error histogram. Each number is a floating number in the
2955 range (0,100], and the maximum length of the list is 20. Use ``:`` to
2956 separate the numbers, and list the numbers in ascending order. For
2957 example, ``--percentile_list=99.5:99.9`` will cause fio to report the
2958 values of completion latency below which 99.5% and 99.9% of the observed
2959 latencies fell, respectively.
f80dba8d 2960
e883cb35
JF
2961.. option:: significant_figures=int
2962
c32ba107
JA
2963 If using :option:`--output-format` of `normal`, set the significant
2964 figures to this value. Higher values will yield more precise IOPS and
2965 throughput units, while lower values will round. Requires a minimum
2966 value of 1 and a maximum value of 10. Defaults to 4.
e883cb35 2967
f80dba8d
MT
2968
2969Error handling
2970~~~~~~~~~~~~~~
2971
2972.. option:: exitall_on_error
2973
2974 When one job finishes in error, terminate the rest. The default is to wait
2975 for each job to finish.
2976
2977.. option:: continue_on_error=str
2978
2979 Normally fio will exit the job on the first observed failure. If this option
2980 is set, fio will continue the job when there is a 'non-fatal error' (EIO or
2981 EILSEQ) until the runtime is exceeded or the I/O size specified is
2982 completed. If this option is used, there are two more stats that are
2983 appended, the total error count and the first error. The error field given
2984 in the stats is the first error that was hit during the run.
2985
2986 The allowed values are:
2987
2988 **none**
2989 Exit on any I/O or verify errors.
2990
2991 **read**
2992 Continue on read errors, exit on all others.
2993
2994 **write**
2995 Continue on write errors, exit on all others.
2996
2997 **io**
2998 Continue on any I/O error, exit on all others.
2999
3000 **verify**
3001 Continue on verify errors, exit on all others.
3002
3003 **all**
3004 Continue on all errors.
3005
3006 **0**
3007 Backward-compatible alias for 'none'.
3008
3009 **1**
3010 Backward-compatible alias for 'all'.
3011
3012.. option:: ignore_error=str
3013
3014 Sometimes you want to ignore some errors during test in that case you can
a35ef7cb
TK
3015 specify error list for each error type, instead of only being able to
3016 ignore the default 'non-fatal error' using :option:`continue_on_error`.
f80dba8d
MT
3017 ``ignore_error=READ_ERR_LIST,WRITE_ERR_LIST,VERIFY_ERR_LIST`` errors for
3018 given error type is separated with ':'. Error may be symbol ('ENOSPC',
3019 'ENOMEM') or integer. Example::
3020
3021 ignore_error=EAGAIN,ENOSPC:122
3022
3023 This option will ignore EAGAIN from READ, and ENOSPC and 122(EDQUOT) from
a35ef7cb
TK
3024 WRITE. This option works by overriding :option:`continue_on_error` with
3025 the list of errors for each error type if any.
f80dba8d
MT
3026
3027.. option:: error_dump=bool
3028
3029 If set dump every error even if it is non fatal, true by default. If
3030 disabled only fatal error will be dumped.
3031
f75ede1d
SW
3032Running predefined workloads
3033----------------------------
3034
3035Fio includes predefined profiles that mimic the I/O workloads generated by
3036other tools.
3037
3038.. option:: profile=str
3039
3040 The predefined workload to run. Current profiles are:
3041
3042 **tiobench**
3043 Threaded I/O bench (tiotest/tiobench) like workload.
3044
3045 **act**
3046 Aerospike Certification Tool (ACT) like workload.
3047
3048To view a profile's additional options use :option:`--cmdhelp` after specifying
3049the profile. For example::
3050
f50fbdda 3051 $ fio --profile=act --cmdhelp
f75ede1d
SW
3052
3053Act profile options
3054~~~~~~~~~~~~~~~~~~~
3055
3056.. option:: device-names=str
3057 :noindex:
3058
3059 Devices to use.
3060
3061.. option:: load=int
3062 :noindex:
3063
3064 ACT load multiplier. Default: 1.
3065
3066.. option:: test-duration=time
3067 :noindex:
3068
947e0fe0
SW
3069 How long the entire test takes to run. When the unit is omitted, the value
3070 is given in seconds. Default: 24h.
f75ede1d
SW
3071
3072.. option:: threads-per-queue=int
3073 :noindex:
3074
f50fbdda 3075 Number of read I/O threads per device. Default: 8.
f75ede1d
SW
3076
3077.. option:: read-req-num-512-blocks=int
3078 :noindex:
3079
3080 Number of 512B blocks to read at the time. Default: 3.
3081
3082.. option:: large-block-op-kbytes=int
3083 :noindex:
3084
3085 Size of large block ops in KiB (writes). Default: 131072.
3086
3087.. option:: prep
3088 :noindex:
3089
3090 Set to run ACT prep phase.
3091
3092Tiobench profile options
3093~~~~~~~~~~~~~~~~~~~~~~~~
3094
3095.. option:: size=str
3096 :noindex:
3097
f50fbdda 3098 Size in MiB.
f75ede1d
SW
3099
3100.. option:: block=int
3101 :noindex:
3102
3103 Block size in bytes. Default: 4096.
3104
3105.. option:: numruns=int
3106 :noindex:
3107
3108 Number of runs.
3109
3110.. option:: dir=str
3111 :noindex:
3112
3113 Test directory.
3114
3115.. option:: threads=int
3116 :noindex:
3117
3118 Number of threads.
f80dba8d
MT
3119
3120Interpreting the output
3121-----------------------
3122
36214730
SW
3123..
3124 Example output was based on the following:
3125 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --time_based \
3126 --rate=1256k --bs=14K --name=quick --runtime=1s --name=mixed \
3127 --runtime=2m --rw=rw
3128
f80dba8d
MT
3129Fio spits out a lot of output. While running, fio will display the status of the
3130jobs created. An example of that would be::
3131
9d25d068 3132 Jobs: 1 (f=1): [_(1),M(1)][24.8%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 01m:31s]
f80dba8d 3133
36214730
SW
3134The characters inside the first set of square brackets denote the current status of
3135each thread. The first character is the first job defined in the job file, and so
3136forth. The possible values (in typical life cycle order) are:
f80dba8d
MT
3137
3138+------+-----+-----------------------------------------------------------+
3139| Idle | Run | |
3140+======+=====+===========================================================+
3141| P | | Thread setup, but not started. |
3142+------+-----+-----------------------------------------------------------+
3143| C | | Thread created. |
3144+------+-----+-----------------------------------------------------------+
3145| I | | Thread initialized, waiting or generating necessary data. |
3146+------+-----+-----------------------------------------------------------+
3147| | p | Thread running pre-reading file(s). |
3148+------+-----+-----------------------------------------------------------+
36214730
SW
3149| | / | Thread is in ramp period. |
3150+------+-----+-----------------------------------------------------------+
f80dba8d
MT
3151| | R | Running, doing sequential reads. |
3152+------+-----+-----------------------------------------------------------+
3153| | r | Running, doing random reads. |
3154+------+-----+-----------------------------------------------------------+
3155| | W | Running, doing sequential writes. |
3156+------+-----+-----------------------------------------------------------+
3157| | w | Running, doing random writes. |
3158+------+-----+-----------------------------------------------------------+
3159| | M | Running, doing mixed sequential reads/writes. |
3160+------+-----+-----------------------------------------------------------+
3161| | m | Running, doing mixed random reads/writes. |
3162+------+-----+-----------------------------------------------------------+
36214730
SW
3163| | D | Running, doing sequential trims. |
3164+------+-----+-----------------------------------------------------------+
3165| | d | Running, doing random trims. |
3166+------+-----+-----------------------------------------------------------+
3167| | F | Running, currently waiting for :manpage:`fsync(2)`. |
f80dba8d
MT
3168+------+-----+-----------------------------------------------------------+
3169| | V | Running, doing verification of written data. |
3170+------+-----+-----------------------------------------------------------+
36214730
SW
3171| f | | Thread finishing. |
3172+------+-----+-----------------------------------------------------------+
f80dba8d
MT
3173| E | | Thread exited, not reaped by main thread yet. |
3174+------+-----+-----------------------------------------------------------+
36214730 3175| _ | | Thread reaped. |
f80dba8d
MT
3176+------+-----+-----------------------------------------------------------+
3177| X | | Thread reaped, exited with an error. |
3178+------+-----+-----------------------------------------------------------+
3179| K | | Thread reaped, exited due to signal. |
3180+------+-----+-----------------------------------------------------------+
3181
36214730
SW
3182..
3183 Example output was based on the following:
3184 TZ=UTC fio --iodepth=8 --ioengine=null --size=100M --runtime=58m \
3185 --time_based --rate=2512k --bs=256K --numjobs=10 \
3186 --name=readers --rw=read --name=writers --rw=write
3187
f80dba8d 3188Fio will condense the thread string as not to take up more space on the command
36214730 3189line than needed. For instance, if you have 10 readers and 10 writers running,
f80dba8d
MT
3190the output would look like this::
3191
9d25d068 3192 Jobs: 20 (f=20): [R(10),W(10)][4.0%][r=20.5MiB/s,w=23.5MiB/s][r=82,w=94 IOPS][eta 57m:36s]
f80dba8d 3193
36214730
SW
3194Note that the status string is displayed in order, so it's possible to tell which of
3195the jobs are currently doing what. In the example above this means that jobs 1--10
3196are readers and 11--20 are writers.
f80dba8d
MT
3197
3198The other values are fairly self explanatory -- number of threads currently
36214730
SW
3199running and doing I/O, the number of currently open files (f=), the estimated
3200completion percentage, the rate of I/O since last check (read speed listed first,
f50fbdda
TK
3201then write speed and optionally trim speed) in terms of bandwidth and IOPS,
3202and time to completion for the current running group. It's impossible to estimate
3203runtime of the following groups (if any).
36214730
SW
3204
3205..
3206 Example output was based on the following:
3207 TZ=UTC fio --iodepth=16 --ioengine=posixaio --filename=/tmp/fiofile \
3208 --direct=1 --size=100M --time_based --runtime=50s --rate_iops=89 \
3209 --bs=7K --name=Client1 --rw=write
3210
3211When fio is done (or interrupted by :kbd:`Ctrl-C`), it will show the data for
3212each thread, group of threads, and disks in that order. For each overall thread (or
3213group) the output looks like::
3214
3215 Client1: (groupid=0, jobs=1): err= 0: pid=16109: Sat Jun 24 12:07:54 2017
3216 write: IOPS=88, BW=623KiB/s (638kB/s)(30.4MiB/50032msec)
3217 slat (nsec): min=500, max=145500, avg=8318.00, stdev=4781.50
3218 clat (usec): min=170, max=78367, avg=4019.02, stdev=8293.31
3219 lat (usec): min=174, max=78375, avg=4027.34, stdev=8291.79
3220 clat percentiles (usec):
3221 | 1.00th=[ 302], 5.00th=[ 326], 10.00th=[ 343], 20.00th=[ 363],
3222 | 30.00th=[ 392], 40.00th=[ 404], 50.00th=[ 416], 60.00th=[ 445],
3223 | 70.00th=[ 816], 80.00th=[ 6718], 90.00th=[12911], 95.00th=[21627],
3224 | 99.00th=[43779], 99.50th=[51643], 99.90th=[68682], 99.95th=[72877],
3225 | 99.99th=[78119]
3226 bw ( KiB/s): min= 532, max= 686, per=0.10%, avg=622.87, stdev=24.82, samples= 100
3227 iops : min= 76, max= 98, avg=88.98, stdev= 3.54, samples= 100
29092211
VF
3228 lat (usec) : 250=0.04%, 500=64.11%, 750=4.81%, 1000=2.79%
3229 lat (msec) : 2=4.16%, 4=1.84%, 10=4.90%, 20=11.33%, 50=5.37%
3230 lat (msec) : 100=0.65%
36214730
SW
3231 cpu : usr=0.27%, sys=0.18%, ctx=12072, majf=0, minf=21
3232 IO depths : 1=85.0%, 2=13.1%, 4=1.8%, 8=0.1%, 16=0.0%, 32=0.0%, >=64=0.0%
3233 submit : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3234 complete : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
3235 issued rwt: total=0,4450,0, short=0,0,0, dropped=0,0,0
3236 latency : target=0, window=0, percentile=100.00%, depth=8
3237
3238The job name (or first job's name when using :option:`group_reporting`) is printed,
3239along with the group id, count of jobs being aggregated, last error id seen (which
3240is 0 when there are no errors), pid/tid of that thread and the time the job/group
3241completed. Below are the I/O statistics for each data direction performed (showing
3242writes in the example above). In the order listed, they denote:
3243
3244**read/write/trim**
3245 The string before the colon shows the I/O direction the statistics
3246 are for. **IOPS** is the average I/Os performed per second. **BW**
3247 is the average bandwidth rate shown as: value in power of 2 format
3248 (value in power of 10 format). The last two values show: (**total
3249 I/O performed** in power of 2 format / **runtime** of that thread).
f80dba8d
MT
3250
3251**slat**
36214730
SW
3252 Submission latency (**min** being the minimum, **max** being the
3253 maximum, **avg** being the average, **stdev** being the standard
3254 deviation). This is the time it took to submit the I/O. For
3255 sync I/O this row is not displayed as the slat is really the
3256 completion latency (since queue/complete is one operation there).
3257 This value can be in nanoseconds, microseconds or milliseconds ---
3258 fio will choose the most appropriate base and print that (in the
3259 example above nanoseconds was the best scale). Note: in :option:`--minimal` mode
0d237712 3260 latencies are always expressed in microseconds.
f80dba8d
MT
3261
3262**clat**
3263 Completion latency. Same names as slat, this denotes the time from
3264 submission to completion of the I/O pieces. For sync I/O, clat will
3265 usually be equal (or very close) to 0, as the time from submit to
3266 complete is basically just CPU time (I/O has already been done, see slat
3267 explanation).
3268
29092211
VF
3269**lat**
3270 Total latency. Same names as slat and clat, this denotes the time from
3271 when fio created the I/O unit to completion of the I/O operation.
3272
f80dba8d 3273**bw**
36214730
SW
3274 Bandwidth statistics based on samples. Same names as the xlat stats,
3275 but also includes the number of samples taken (**samples**) and an
3276 approximate percentage of total aggregate bandwidth this thread
3277 received in its group (**per**). This last value is only really
3278 useful if the threads in this group are on the same disk, since they
3279 are then competing for disk access.
3280
3281**iops**
3282 IOPS statistics based on samples. Same names as bw.
f80dba8d 3283
29092211
VF
3284**lat (nsec/usec/msec)**
3285 The distribution of I/O completion latencies. This is the time from when
3286 I/O leaves fio and when it gets completed. Unlike the separate
3287 read/write/trim sections above, the data here and in the remaining
3288 sections apply to all I/Os for the reporting group. 250=0.04% means that
3289 0.04% of the I/Os completed in under 250us. 500=64.11% means that 64.11%
3290 of the I/Os required 250 to 499us for completion.
3291
f80dba8d
MT
3292**cpu**
3293 CPU usage. User and system time, along with the number of context
3294 switches this thread went through, usage of system and user time, and
3295 finally the number of major and minor page faults. The CPU utilization
3296 numbers are averages for the jobs in that reporting group, while the
23a8e176 3297 context and fault counters are summed.
f80dba8d
MT
3298
3299**IO depths**
a2140525
SW
3300 The distribution of I/O depths over the job lifetime. The numbers are
3301 divided into powers of 2 and each entry covers depths from that value
3302 up to those that are lower than the next entry -- e.g., 16= covers
3303 depths from 16 to 31. Note that the range covered by a depth
3304 distribution entry can be different to the range covered by the
3305 equivalent submit/complete distribution entry.
f80dba8d
MT
3306
3307**IO submit**
3308 How many pieces of I/O were submitting in a single submit call. Each
c60ebc45 3309 entry denotes that amount and below, until the previous entry -- e.g.,
a2140525
SW
3310 16=100% means that we submitted anywhere between 9 to 16 I/Os per submit
3311 call. Note that the range covered by a submit distribution entry can
3312 be different to the range covered by the equivalent depth distribution
3313 entry.
f80dba8d
MT
3314
3315**IO complete**
3316 Like the above submit number, but for completions instead.
3317
36214730
SW
3318**IO issued rwt**
3319 The number of read/write/trim requests issued, and how many of them were
3320 short or dropped.
f80dba8d 3321
29092211 3322**IO latency**
ee21ebee 3323 These values are for :option:`latency_target` and related options. When
29092211
VF
3324 these options are engaged, this section describes the I/O depth required
3325 to meet the specified latency target.
71bfa161 3326
36214730
SW
3327..
3328 Example output was based on the following:
3329 TZ=UTC fio --ioengine=null --iodepth=2 --size=100M --numjobs=2 \
3330 --rate_process=poisson --io_limit=32M --name=read --bs=128k \
3331 --rate=11M --name=write --rw=write --bs=2k --rate=700k
3332
71bfa161 3333After each client has been listed, the group statistics are printed. They
f80dba8d 3334will look like this::
71bfa161 3335
f80dba8d 3336 Run status group 0 (all jobs):
36214730
SW
3337 READ: bw=20.9MiB/s (21.9MB/s), 10.4MiB/s-10.8MiB/s (10.9MB/s-11.3MB/s), io=64.0MiB (67.1MB), run=2973-3069msec
3338 WRITE: bw=1231KiB/s (1261kB/s), 616KiB/s-621KiB/s (630kB/s-636kB/s), io=64.0MiB (67.1MB), run=52747-53223msec
71bfa161 3339
36214730 3340For each data direction it prints:
71bfa161 3341
36214730
SW
3342**bw**
3343 Aggregate bandwidth of threads in this group followed by the
3344 minimum and maximum bandwidth of all the threads in this group.
3345 Values outside of brackets are power-of-2 format and those
3346 within are the equivalent value in a power-of-10 format.
f80dba8d 3347**io**
36214730
SW
3348 Aggregate I/O performed of all threads in this group. The
3349 format is the same as bw.
3350**run**
3351 The smallest and longest runtimes of the threads in this group.
71bfa161 3352
f50fbdda 3353And finally, the disk statistics are printed. This is Linux specific. They will look like this::
71bfa161 3354
f80dba8d
MT
3355 Disk stats (read/write):
3356 sda: ios=16398/16511, merge=30/162, ticks=6853/819634, in_queue=826487, util=100.00%
71bfa161
JA
3357
3358Each value is printed for both reads and writes, with reads first. The
3359numbers denote:
3360
f80dba8d 3361**ios**
c60ebc45 3362 Number of I/Os performed by all groups.
f80dba8d 3363**merge**
007c7be9 3364 Number of merges performed by the I/O scheduler.
f80dba8d
MT
3365**ticks**
3366 Number of ticks we kept the disk busy.
36214730 3367**in_queue**
f80dba8d
MT
3368 Total time spent in the disk queue.
3369**util**
3370 The disk utilization. A value of 100% means we kept the disk
71bfa161
JA
3371 busy constantly, 50% would be a disk idling half of the time.
3372
f80dba8d
MT
3373It is also possible to get fio to dump the current output while it is running,
3374without terminating the job. To do that, send fio the **USR1** signal. You can
3375also get regularly timed dumps by using the :option:`--status-interval`
3376parameter, or by creating a file in :file:`/tmp` named
3377:file:`fio-dump-status`. If fio sees this file, it will unlink it and dump the
3378current output status.
8423bd11 3379
71bfa161 3380
f80dba8d
MT
3381Terse output
3382------------
71bfa161 3383
f80dba8d
MT
3384For scripted usage where you typically want to generate tables or graphs of the
3385results, fio can output the results in a semicolon separated format. The format
3386is one long line of values, such as::
71bfa161 3387
f80dba8d
MT
3388 2;card0;0;0;7139336;121836;60004;1;10109;27.932460;116.933948;220;126861;3495.446807;1085.368601;226;126864;3523.635629;1089.012448;24063;99944;50.275485%;59818.274627;5540.657370;7155060;122104;60004;1;8338;29.086342;117.839068;388;128077;5032.488518;1234.785715;391;128085;5061.839412;1236.909129;23436;100928;50.287926%;59964.832030;5644.844189;14.595833%;19.394167%;123706;0;7313;0.1%;0.1%;0.1%;0.1%;0.1%;0.1%;100.0%;0.00%;0.00%;0.00%;0.00%;0.00%;0.00%;0.01%;0.02%;0.05%;0.16%;6.04%;40.40%;52.68%;0.64%;0.01%;0.00%;0.01%;0.00%;0.00%;0.00%;0.00%;0.00%
3389 A description of this job goes here.
562c2d2f
DN
3390
3391The job description (if provided) follows on a second line.
71bfa161 3392
a7f77fa6
SW
3393To enable terse output, use the :option:`--minimal` or
3394:option:`--output-format`\=terse command line options. The
f80dba8d
MT
3395first value is the version of the terse output format. If the output has to be
3396changed for some reason, this number will be incremented by 1 to signify that
3397change.
6820cb3b 3398
a2c95580 3399Split up, the format is as follows (comments in brackets denote when a
007c7be9 3400field was introduced or whether it's specific to some terse version):
71bfa161 3401
f80dba8d
MT
3402 ::
3403
f50fbdda 3404 terse version, fio version [v3], jobname, groupid, error
f80dba8d
MT
3405
3406 READ status::
3407
3408 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3409 Submission latency: min, max, mean, stdev (usec)
3410 Completion latency: min, max, mean, stdev (usec)
3411 Completion latency percentiles: 20 fields (see below)
3412 Total latency: min, max, mean, stdev (usec)
f50fbdda
TK
3413 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3414 IOPS [v5]: min, max, mean, stdev, number of samples
f80dba8d
MT
3415
3416 WRITE status:
3417
3418 ::
3419
3420 Total IO (KiB), bandwidth (KiB/sec), IOPS, runtime (msec)
3421 Submission latency: min, max, mean, stdev (usec)
247823cc 3422 Completion latency: min, max, mean, stdev (usec)
f80dba8d
MT
3423 Completion latency percentiles: 20 fields (see below)
3424 Total latency: min, max, mean, stdev (usec)
f50fbdda
TK
3425 Bw (KiB/s): min, max, aggregate percentage of total, mean, stdev, number of samples [v5]
3426 IOPS [v5]: min, max, mean, stdev, number of samples
a2c95580
AH
3427
3428 TRIM status [all but version 3]:
3429
f50fbdda 3430 Fields are similar to READ/WRITE status.
f80dba8d
MT
3431
3432 CPU usage::
3433
3434 user, system, context switches, major faults, minor faults
3435
3436 I/O depths::
3437
3438 <=1, 2, 4, 8, 16, 32, >=64
3439
3440 I/O latencies microseconds::
3441
3442 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000
3443
3444 I/O latencies milliseconds::
3445
3446 <=2, 4, 10, 20, 50, 100, 250, 500, 750, 1000, 2000, >=2000
3447
a2c95580 3448 Disk utilization [v3]::
f80dba8d 3449
f50fbdda
TK
3450 disk name, read ios, write ios, read merges, write merges, read ticks, write ticks,
3451 time spent in queue, disk utilization percentage
f80dba8d
MT
3452
3453 Additional Info (dependent on continue_on_error, default off)::
3454
3455 total # errors, first error code
3456
3457 Additional Info (dependent on description being set)::
3458
3459 Text description
3460
3461Completion latency percentiles can be a grouping of up to 20 sets, so for the
3462terse output fio writes all of them. Each field will look like this::
1db92cb6 3463
f50fbdda 3464 1.00%=6112
1db92cb6 3465
f80dba8d 3466which is the Xth percentile, and the `usec` latency associated with it.
1db92cb6 3467
f50fbdda 3468For `Disk utilization`, all disks used by fio are shown. So for each disk there
f80dba8d 3469will be a disk utilization section.
f2f788dd 3470
2fc26c3d 3471Below is a single line containing short names for each of the fields in the
2831be97 3472minimal output v3, separated by semicolons::
2fc26c3d 3473
f50fbdda 3474 terse_version_3;fio_version;jobname;groupid;error;read_kb;read_bandwidth;read_iops;read_runtime_ms;read_slat_min;read_slat_max;read_slat_mean;read_slat_dev;read_clat_min;read_clat_max;read_clat_mean;read_clat_dev;read_clat_pct01;read_clat_pct02;read_clat_pct03;read_clat_pct04;read_clat_pct05;read_clat_pct06;read_clat_pct07;read_clat_pct08;read_clat_pct09;read_clat_pct10;read_clat_pct11;read_clat_pct12;read_clat_pct13;read_clat_pct14;read_clat_pct15;read_clat_pct16;read_clat_pct17;read_clat_pct18;read_clat_pct19;read_clat_pct20;read_tlat_min;read_lat_max;read_lat_mean;read_lat_dev;read_bw_min;read_bw_max;read_bw_agg_pct;read_bw_mean;read_bw_dev;write_kb;write_bandwidth;write_iops;write_runtime_ms;write_slat_min;write_slat_max;write_slat_mean;write_slat_dev;write_clat_min;write_clat_max;write_clat_mean;write_clat_dev;write_clat_pct01;write_clat_pct02;write_clat_pct03;write_clat_pct04;write_clat_pct05;write_clat_pct06;write_clat_pct07;write_clat_pct08;write_clat_pct09;write_clat_pct10;write_clat_pct11;write_clat_pct12;write_clat_pct13;write_clat_pct14;write_clat_pct15;write_clat_pct16;write_clat_pct17;write_clat_pct18;write_clat_pct19;write_clat_pct20;write_tlat_min;write_lat_max;write_lat_mean;write_lat_dev;write_bw_min;write_bw_max;write_bw_agg_pct;write_bw_mean;write_bw_dev;cpu_user;cpu_sys;cpu_csw;cpu_mjf;cpu_minf;iodepth_1;iodepth_2;iodepth_4;iodepth_8;iodepth_16;iodepth_32;iodepth_64;lat_2us;lat_4us;lat_10us;lat_20us;lat_50us;lat_100us;lat_250us;lat_500us;lat_750us;lat_1000us;lat_2ms;lat_4ms;lat_10ms;lat_20ms;lat_50ms;lat_100ms;lat_250ms;lat_500ms;lat_750ms;lat_1000ms;lat_2000ms;lat_over_2000ms;disk_name;disk_read_iops;disk_write_iops;disk_read_merges;disk_write_merges;disk_read_ticks;write_ticks;disk_queue_time;disk_util
2fc26c3d 3475
25c8b9d7 3476
44c82dba
VF
3477JSON output
3478------------
3479
3480The `json` output format is intended to be both human readable and convenient
3481for automated parsing. For the most part its sections mirror those of the
3482`normal` output. The `runtime` value is reported in msec and the `bw` value is
3483reported in 1024 bytes per second units.
3484
3485
d29c4a91
VF
3486JSON+ output
3487------------
3488
3489The `json+` output format is identical to the `json` output format except that it
3490adds a full dump of the completion latency bins. Each `bins` object contains a
3491set of (key, value) pairs where keys are latency durations and values count how
3492many I/Os had completion latencies of the corresponding duration. For example,
3493consider:
3494
3495 "bins" : { "87552" : 1, "89600" : 1, "94720" : 1, "96768" : 1, "97792" : 1, "99840" : 1, "100864" : 2, "103936" : 6, "104960" : 534, "105984" : 5995, "107008" : 7529, ... }
3496
3497This data indicates that one I/O required 87,552ns to complete, two I/Os required
3498100,864ns to complete, and 7529 I/Os required 107,008ns to complete.
3499
3500Also included with fio is a Python script `fio_jsonplus_clat2csv` that takes
3501json+ output and generates CSV-formatted latency data suitable for plotting.
3502
3503The latency durations actually represent the midpoints of latency intervals.
f50fbdda 3504For details refer to :file:`stat.h`.
d29c4a91
VF
3505
3506
f80dba8d
MT
3507Trace file format
3508-----------------
3509
3510There are two trace file format that you can encounter. The older (v1) format is
3511unsupported since version 1.20-rc3 (March 2008). It will still be described
25c8b9d7
PD
3512below in case that you get an old trace and want to understand it.
3513
3514In any case the trace is a simple text file with a single action per line.
3515
3516
f80dba8d
MT
3517Trace file format v1
3518~~~~~~~~~~~~~~~~~~~~
3519
3520Each line represents a single I/O action in the following format::
3521
3522 rw, offset, length
25c8b9d7 3523
f50fbdda 3524where `rw=0/1` for read/write, and the `offset` and `length` entries being in bytes.
25c8b9d7 3525
22413915 3526This format is not supported in fio versions >= 1.20-rc3.
25c8b9d7 3527
25c8b9d7 3528
f80dba8d
MT
3529Trace file format v2
3530~~~~~~~~~~~~~~~~~~~~
25c8b9d7 3531
f80dba8d
MT
3532The second version of the trace file format was added in fio version 1.17. It
3533allows to access more then one file per trace and has a bigger set of possible
3534file actions.
25c8b9d7 3535
f80dba8d 3536The first line of the trace file has to be::
25c8b9d7 3537
f80dba8d 3538 fio version 2 iolog
25c8b9d7
PD
3539
3540Following this can be lines in two different formats, which are described below.
3541
f80dba8d 3542The file management format::
25c8b9d7 3543
f80dba8d 3544 filename action
25c8b9d7 3545
f50fbdda 3546The `filename` is given as an absolute path. The `action` can be one of these:
25c8b9d7 3547
f80dba8d 3548**add**
f50fbdda 3549 Add the given `filename` to the trace.
f80dba8d 3550**open**
f50fbdda 3551 Open the file with the given `filename`. The `filename` has to have
f80dba8d
MT
3552 been added with the **add** action before.
3553**close**
f50fbdda 3554 Close the file with the given `filename`. The file has to have been
f80dba8d
MT
3555 opened before.
3556
3557
3558The file I/O action format::
3559
3560 filename action offset length
3561
3562The `filename` is given as an absolute path, and has to have been added and
3563opened before it can be used with this format. The `offset` and `length` are
3564given in bytes. The `action` can be one of these:
3565
3566**wait**
3567 Wait for `offset` microseconds. Everything below 100 is discarded.
3568 The time is relative to the previous `wait` statement.
3569**read**
3570 Read `length` bytes beginning from `offset`.
3571**write**
3572 Write `length` bytes beginning from `offset`.
3573**sync**
3574 :manpage:`fsync(2)` the file.
3575**datasync**
3576 :manpage:`fdatasync(2)` the file.
3577**trim**
3578 Trim the given file from the given `offset` for `length` bytes.
3579
3580CPU idleness profiling
3581----------------------
3582
3583In some cases, we want to understand CPU overhead in a test. For example, we
3584test patches for the specific goodness of whether they reduce CPU usage.
3585Fio implements a balloon approach to create a thread per CPU that runs at idle
3586priority, meaning that it only runs when nobody else needs the cpu.
3587By measuring the amount of work completed by the thread, idleness of each CPU
3588can be derived accordingly.
3589
3590An unit work is defined as touching a full page of unsigned characters. Mean and
3591standard deviation of time to complete an unit work is reported in "unit work"
3592section. Options can be chosen to report detailed percpu idleness or overall
3593system idleness by aggregating percpu stats.
3594
3595
3596Verification and triggers
3597-------------------------
3598
3599Fio is usually run in one of two ways, when data verification is done. The first
3600is a normal write job of some sort with verify enabled. When the write phase has
3601completed, fio switches to reads and verifies everything it wrote. The second
3602model is running just the write phase, and then later on running the same job
3603(but with reads instead of writes) to repeat the same I/O patterns and verify
3604the contents. Both of these methods depend on the write phase being completed,
3605as fio otherwise has no idea how much data was written.
3606
3607With verification triggers, fio supports dumping the current write state to
3608local files. Then a subsequent read verify workload can load this state and know
3609exactly where to stop. This is useful for testing cases where power is cut to a
3610server in a managed fashion, for instance.
99b9a85a
JA
3611
3612A verification trigger consists of two things:
3613
f80dba8d
MT
36141) Storing the write state of each job.
36152) Executing a trigger command.
99b9a85a 3616
f80dba8d
MT
3617The write state is relatively small, on the order of hundreds of bytes to single
3618kilobytes. It contains information on the number of completions done, the last X
3619completions, etc.
99b9a85a 3620
f80dba8d
MT
3621A trigger is invoked either through creation ('touch') of a specified file in
3622the system, or through a timeout setting. If fio is run with
9207a0cb 3623:option:`--trigger-file`\= :file:`/tmp/trigger-file`, then it will continually
f80dba8d
MT
3624check for the existence of :file:`/tmp/trigger-file`. When it sees this file, it
3625will fire off the trigger (thus saving state, and executing the trigger
99b9a85a
JA
3626command).
3627
f80dba8d
MT
3628For client/server runs, there's both a local and remote trigger. If fio is
3629running as a server backend, it will send the job states back to the client for
3630safe storage, then execute the remote trigger, if specified. If a local trigger
3631is specified, the server will still send back the write state, but the client
3632will then execute the trigger.
99b9a85a 3633
f80dba8d
MT
3634Verification trigger example
3635~~~~~~~~~~~~~~~~~~~~~~~~~~~~
99b9a85a 3636
f50fbdda
TK
3637Let's say we want to run a powercut test on the remote Linux machine 'server'.
3638Our write workload is in :file:`write-test.fio`. We want to cut power to 'server' at
f80dba8d
MT
3639some point during the run, and we'll run this test from the safety or our local
3640machine, 'localbox'. On the server, we'll start the fio backend normally::
99b9a85a 3641
f80dba8d 3642 server# fio --server
99b9a85a 3643
f80dba8d 3644and on the client, we'll fire off the workload::
99b9a85a 3645
f80dba8d 3646 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger-remote="bash -c \"echo b > /proc/sysrq-triger\""
99b9a85a 3647
f80dba8d 3648We set :file:`/tmp/my-trigger` as the trigger file, and we tell fio to execute::
99b9a85a 3649
f80dba8d 3650 echo b > /proc/sysrq-trigger
99b9a85a 3651
f80dba8d
MT
3652on the server once it has received the trigger and sent us the write state. This
3653will work, but it's not **really** cutting power to the server, it's merely
3654abruptly rebooting it. If we have a remote way of cutting power to the server
3655through IPMI or similar, we could do that through a local trigger command
4502cb42 3656instead. Let's assume we have a script that does IPMI reboot of a given hostname,
f80dba8d
MT
3657ipmi-reboot. On localbox, we could then have run fio with a local trigger
3658instead::
99b9a85a 3659
f80dba8d 3660 localbox$ fio --client=server --trigger-file=/tmp/my-trigger --trigger="ipmi-reboot server"
99b9a85a 3661
f80dba8d
MT
3662For this case, fio would wait for the server to send us the write state, then
3663execute ``ipmi-reboot server`` when that happened.
3664
3665Loading verify state
3666~~~~~~~~~~~~~~~~~~~~
3667
4502cb42 3668To load stored write state, a read verification job file must contain the
f80dba8d 3669:option:`verify_state_load` option. If that is set, fio will load the previously
99b9a85a 3670stored state. For a local fio run this is done by loading the files directly,
f80dba8d
MT
3671and on a client/server run, the server backend will ask the client to send the
3672files over and load them from there.
a3ae5b05
JA
3673
3674
f80dba8d
MT
3675Log File Formats
3676----------------
a3ae5b05
JA
3677
3678Fio supports a variety of log file formats, for logging latencies, bandwidth,
3679and IOPS. The logs share a common format, which looks like this:
3680
5a83478f
SW
3681 *time* (`msec`), *value*, *data direction*, *block size* (`bytes`),
3682 *offset* (`bytes`)
a3ae5b05 3683
5a83478f 3684*Time* for the log entry is always in milliseconds. The *value* logged depends
a3ae5b05
JA
3685on the type of log, it will be one of the following:
3686
f80dba8d 3687 **Latency log**
168bb587 3688 Value is latency in nsecs
f80dba8d
MT
3689 **Bandwidth log**
3690 Value is in KiB/sec
3691 **IOPS log**
3692 Value is IOPS
3693
3694*Data direction* is one of the following:
3695
3696 **0**
3697 I/O is a READ
3698 **1**
3699 I/O is a WRITE
3700 **2**
3701 I/O is a TRIM
3702
5a83478f
SW
3703The entry's *block size* is always in bytes. The *offset* is the offset, in bytes,
3704from the start of the file, for that particular I/O. The logging of the offset can be
3705toggled with :option:`log_offset`.
f80dba8d 3706
6fc82095 3707Fio defaults to logging every individual I/O. When IOPS are logged for individual
5a83478f 3708I/Os the *value* entry will always be 1. If windowed logging is enabled through
6fc82095
SW
3709:option:`log_avg_msec`, fio logs the average values over the specified period of time.
3710If windowed logging is enabled and :option:`log_max_value` is set, then fio logs
5a83478f
SW
3711maximum values in that window instead of averages. Since *data direction*, *block
3712size* and *offset* are per-I/O values, if windowed logging is enabled they
3713aren't applicable and will be 0.
f80dba8d 3714
b8f7e412 3715Client/Server
f80dba8d
MT
3716-------------
3717
3718Normally fio is invoked as a stand-alone application on the machine where the
6cf30ac0
SW
3719I/O workload should be generated. However, the backend and frontend of fio can
3720be run separately i.e., the fio server can generate an I/O workload on the "Device
3721Under Test" while being controlled by a client on another machine.
f80dba8d
MT
3722
3723Start the server on the machine which has access to the storage DUT::
3724
f50fbdda 3725 $ fio --server=args
f80dba8d 3726
dbb257bb 3727where `args` defines what fio listens to. The arguments are of the form
f80dba8d
MT
3728``type,hostname`` or ``IP,port``. *type* is either ``ip`` (or ip4) for TCP/IP
3729v4, ``ip6`` for TCP/IP v6, or ``sock`` for a local unix domain socket.
3730*hostname* is either a hostname or IP address, and *port* is the port to listen
3731to (only valid for TCP/IP, not a local socket). Some examples:
3732
37331) ``fio --server``
3734
3735 Start a fio server, listening on all interfaces on the default port (8765).
3736
37372) ``fio --server=ip:hostname,4444``
3738
3739 Start a fio server, listening on IP belonging to hostname and on port 4444.
3740
37413) ``fio --server=ip6:::1,4444``
3742
3743 Start a fio server, listening on IPv6 localhost ::1 and on port 4444.
3744
37454) ``fio --server=,4444``
3746
3747 Start a fio server, listening on all interfaces on port 4444.
3748
37495) ``fio --server=1.2.3.4``
3750
3751 Start a fio server, listening on IP 1.2.3.4 on the default port.
3752
37536) ``fio --server=sock:/tmp/fio.sock``
3754
dbb257bb 3755 Start a fio server, listening on the local socket :file:`/tmp/fio.sock`.
f80dba8d
MT
3756
3757Once a server is running, a "client" can connect to the fio server with::
3758
3759 fio <local-args> --client=<server> <remote-args> <job file(s)>
3760
3761where `local-args` are arguments for the client where it is running, `server`
3762is the connect string, and `remote-args` and `job file(s)` are sent to the
3763server. The `server` string follows the same format as it does on the server
3764side, to allow IP/hostname/socket and port strings.
3765
3766Fio can connect to multiple servers this way::
3767
3768 fio --client=<server1> <job file(s)> --client=<server2> <job file(s)>
3769
3770If the job file is located on the fio server, then you can tell the server to
3771load a local file as well. This is done by using :option:`--remote-config` ::
3772
3773 fio --client=server --remote-config /path/to/file.fio
3774
3775Then fio will open this local (to the server) job file instead of being passed
3776one from the client.
3777
3778If you have many servers (example: 100 VMs/containers), you can input a pathname
3779of a file containing host IPs/names as the parameter value for the
3780:option:`--client` option. For example, here is an example :file:`host.list`
3781file containing 2 hostnames::
3782
3783 host1.your.dns.domain
3784 host2.your.dns.domain
3785
3786The fio command would then be::
a3ae5b05 3787
f80dba8d 3788 fio --client=host.list <job file(s)>
a3ae5b05 3789
f80dba8d
MT
3790In this mode, you cannot input server-specific parameters or job files -- all
3791servers receive the same job file.
a3ae5b05 3792
f80dba8d
MT
3793In order to let ``fio --client`` runs use a shared filesystem from multiple
3794hosts, ``fio --client`` now prepends the IP address of the server to the
4502cb42 3795filename. For example, if fio is using the directory :file:`/mnt/nfs/fio` and is
f80dba8d
MT
3796writing filename :file:`fileio.tmp`, with a :option:`--client` `hostfile`
3797containing two hostnames ``h1`` and ``h2`` with IP addresses 192.168.10.120 and
3798192.168.10.121, then fio will create two files::
a3ae5b05 3799
f80dba8d
MT
3800 /mnt/nfs/fio/192.168.10.120.fileio.tmp